© 2016 Fire Protection Research Foundation

1 Batterymarch Park, Quincy, MA 02169-7417, USA Email: foundation@nfpa.org | Web: nfpa.org/foundation

Pilot Evaluation of the Remembering WhenTM

Program in Five Communities in Iowa FINAL REPORT BY:

Carri Casteel, Ph.D., Rebecca Bruening, and Sato Ashida, Ph.D. University of Iowa Iowa City, IA, USA February 2016

—— Page i ——

—— Page ii ——

FOREWORD Remembering When™ is a fire and fall prevention program for older adults developed by the National Fire Protection Association (NFPA) and the Centers for Disease Control and Prevention (CDC). It is designed to help older adults live safely at home as long as possible. There are sixteen key messages, eight related to fire prevention and eight related to fall prevention, developed by experts from national and local safety organizations as well as through focus group testing in high-fire-risk states. The program is intended to be led by members of the fire service. Remembering When™ offers flexibility to form a coalition of professionals and volunteers who assist with the group presentations and home visit delivery. Coalition partners include service clubs, social and religious organizations, and retirement communities. Coalition members determine the best way to implement program components based on the unique needs of the community. NFPA wanted to evaluate the program to determine the program’s efficacy, as well as identify any areas where the program could be improved.

The Fire Protection Research Foundation initiated this pilot project to evaluate the efficacy of the Remembering When™ program for both fire and fall prevention and develop recommendations for improvements to the program. The Fire Protection Research Foundation expresses gratitude to the report authors Carri Casteel, Ph.D., Rebecca Bruening, M.P.H. and Sato Ashida, Ph.D., who are with the University of Iowa. The Research Foundation appreciates the guidance provided by the Project Technical Panelists and all others that contributed to this research effort. Special thanks are expressed to the National Fire Protection Association (NFPA) for providing the project funding. The content, opinions and conclusions contained in this report are solely those of the authors and do not necessarily represent the views of the Fire Protection Research Foundation, NFPA, Technical Panel or Sponsors. The Foundation makes no guaranty or warranty as to the accuracy or completeness of any information published herein.

—— Page iii ——

About the Fire Protection Research Foundation

The Fire Protection Research Foundation plans, manages, and communicates research on a broad range of fire safety issues in collaboration with scientists and laboratories around the world. The Foundation is an affiliate of NFPA.

About the National Fire Protection Association (NFPA)

Founded in 1896, NFPA is a global, nonprofit organization devoted to eliminating death, injury, property and economic loss due to fire, electrical and related hazards. The association delivers information and knowledge through more than 300 consensus codes and standards, research, training, education, outreach and advocacy; and by partnering with others who share an interest in furthering the NFPA mission. All NFPA codes and standards can be viewed online for free. NFPA's membership totals more than 65,000 individuals around the world. Keywords: Remembering When, fire prevention, fall prevention, older adults, program efficacy

—— Page iv ——

TECHNICAL PANEL

Beverly Bolton, North Las Vegas Fire Department (retired)

Virginia Charter, Oklahoma State University

Eileen Dryden, Ph.D., Institute for Community Health

Timo Juurakko, Maple Ridge Fire Department, British Columbia, Canada

Leigh Kish, Charlotte Fire Department

Walter Palmer, Ph.D.

Bernadette Wright, Ph.D., Meaningful Evidence LLC

Christina Holcroft, Ph.D., NFPA

Karen Berard-Reed, NFPA

PROJECT SPONSOR

National Fire Protection Association

—— Page v ——

—— Page vi ——

Table of Contents Executive Summary ....................................................................................................................................... 1

Background ............................................................................................................................................... 1

Methods .................................................................................................................................................... 1

Findings ..................................................................................................................................................... 2

Project Background ....................................................................................................................................... 4

Methodology ................................................................................................................................................. 6

1. Overview ............................................................................................................................................... 6

2. Outreach Teams ................................................................................................................................... 6

2.a. Recruitment of Outreach Teams .................................................................................................... 6

2.b. Remembering When™ Program Delivery ..................................................................................... 6

2.c. Exit Interview .................................................................................................................................. 7

3. Older Adult Participants ........................................................................................................................ 7

3.a. Recruitment and Randomization ................................................................................................... 7

3.b. Data Collection ............................................................................................................................... 9

3.c. Measurements ............................................................................................................................. 11

3.d. Data analysis ................................................................................................................................ 15

Results ......................................................................................................................................................... 16

1. Remembering When™ Program Delivery ........................................................................................... 16

2. Older Adult Study Population ............................................................................................................. 16

2.a. Demographics, Housing and Health Characteristics of Enrolled Study Population (n=130)………16

2.b. Demographics and Fall and Fire Risks at Baseline among Sample with Randomly Selected Cohabitating Individual (n=110), Overall and by Study Arm…………………………………………………….17

2.c. Fall and Fire Prevention Behaviors among Sample with Randomly Selected Cohabitating Individual (n=110)……………………………………………………………………………………………………………………20

3. Remembering When™ Program Efficacy ............................................................................................ 21

3.a. Remembering WhenTM Safety Messages: Falls ............................................................................ 21

3.b. Remembering WhenTM Safety Messages: Fires ........................................................................... 22

3.c. Perceived Susceptibility, Severity, and Fear of Falls and Fire ...................................................... 29

3.d. Perceptions about Prevention Behaviors for Falls and Fire ......................................................... 29

3.e. Perceptions about Social Support from Others ........................................................................... 30

3.f. Discussion about Falls and Fire Prevention with Others .............................................................. 33

—— Page vii ——

4. Program Satisfaction and Suggestions for Improvement ................................................................... 33

4.a. Older Adult Participant Feedback ................................................................................................ 33

4.b. Outreach Team Member Feedback ............................................................................................. 34

Discussion.................................................................................................................................................... 42

Recommendations ...................................................................................................................................... 44

Acknowledgments ....................................................................................................................................... 46

References .................................................................................................................................................. 47

—— Page 1 ——

Executive Summary

Background Falls are the leading cause of unintentional fatal and nonfatal injury among older adults in the United States. More than one-third of older adults fall each year, and many of these falls result in decreased or permanent loss of functioning and a loss of independence. Older adults also have one of the highest rates of death from house fires, compared to their younger counterparts. Older adults are more likely to suffer from reduced sensory abilities and mental health, which reduces reaction time and puts them at higher risk for causing a fire and dying. Remembering WhenTM: A Fire and Fall Prevention Program for Older Adults was developed by the National Fire Protection Association (NFPA) and Centers for Disease Control and Prevention to increase knowledge about how older adults can reduce their risk of experiencing a fall and house fire. The program has 16 falls and fire prevention messages, primarily targeting residential hazard and safety behaviors, and can be delivered in both group presentation and home visit formats. In 2015, the University of Iowa partnered with the NFPA and five fire departments in Iowa to evaluate the effectiveness of the Remembering When™ program in: (1) Improving perceptions associated with falls and house fires, and (2) Changing falls prevention and fire safety behaviors.

Methods The Remembering When™ Pilot Evaluation Study was conducted between February and November, 2015 in five communities across Iowa. The five study sites were selected based on local fire department completion of Remembering When™ training and interest in participating in the study. Within each community, the research team collaborated with up to two fire department officials comprising an Outreach Team, which was primarily responsible for delivering the Remembering When™ program and providing data on delivery activities. In addition, adults ages 65 and older were recruited in each community to participate in the study by the research staff. Older adults were primarily recruited from senior residential complexes (not including assisted living or nursing home complexes), home visiting and meal delivery organizations, congregate meal sites, and places of worship.

Eligible older adults were randomly assigned to one of two program delivery methods: (1) those receiving Remembering When™ through a home visit, and (2) those receiving Remembering When™ through both a group presentation and home visit. Each participant completed three telephone interviews and participated in the Remembering WhenTM program training provided by their fire departments. A total of 149 participants enrolled in the study and completed the first interview, of which 133 attended program training conducted in their homes and/or in a group presentation. At the end of the study, 130 of the 133 participants receiving the program completed the final interview to report whether their fire and falls prevention behaviors, and their perceptions associated with fires and

falls, changed after participating in the Remembering WhenTM program.

—— Page 2 ——

Findings • Program Delivery: Overall, the Outreach Teams delivered the Remembering When™ key messages

with fidelity. There was variation in the extent to which Outreach Team members covered each of the key fall and fire safety messages and used supportive materials. However, some of this variation is likely to have been due to the different needs of the participants.

• Older Adult Participant Characteristics: More than half of the participants (66%) were at least 75 years of age, and the majority were female (76%) and White (98%). Sixty percent of the participants lived in an apartment or condominium, and approximately one-third (34%) lived in a house. More than two-thirds of the participants reported an annual income below $50,000. Most of the participants were non-smokers and/or lived in a household without a smoker (95%). Just over half of the participants (55%) reported being in very good or excellent health, while another 35% reported being in good health.

• Outreach Team Characteristics: Ten individuals from five Iowa fire departments participated in the study. Most had been working with community service programs and with older adults for more than ten years. One fire department official had experience delivering the Remembering When™ program prior to the study.

• Risk Factors for Falls and Fires: Older adult participants exhibited risk factors for falls, including older age, female gender and overall multifactorial risk based on a validated fall risk self-assessment. They also exhibited risk factors for experiencing a house fire, including older age, lower household income, and rental residential tenure. We found no differences in fall and fire risks between participants in the home visit only study arm and participants in the home visit and group presentation study arm with respect to demographics and fall and fire risks.

• Participation in Falls and Fire Safety Programs: Ten percent of the participants reported ever having participated in a falls prevention program, and 18% reported ever having participated in a fire safety program prior to the Remembering WhenTM program.

• Changes in Falls Prevention Behaviors: Prior to receiving the program, most participants were engaged in program-recommended activities, including exercising regularly (86%), taking their time to get up from sitting or lying down (83%), keeping stairs and walking areas free of clutter (73%), and keeping a well-lit path between the bedroom and bathroom at night (73%). Following the Remembering WhenTM program, more participants cleared their paths of travel (baseline: n=104, 95%; follow-up: n=80, 73%), used non-slip mats (baseline: n=72, 66%; follow-up: n=86, 78%), turned on lights before using the stairs (baseline: n=54, 49%; follow-up: n=51, 46%), and installed grab bars on the walls next to the bathtub, shower and/or toilet (baseline: n=60, 55%; follow-up: n=78, 71%). We did not observe differences in the percentage of participants improving their falls prevention practices between the two study arms.

• Changes in Fire Prevention Behaviors: Most of the participants were engaged in fire safety practices prior to receiving the Remembering WhenTM program, including having a smoke alarm installed

—— Page 3 ——

outside each sleeping area (91%), staying in the kitchen when frying foods (85%), having a telephone near their beds (79%), having a smoke alarm on every level of their home (78%) and inside each bedroom (76%), and keeping space heaters at least 3 feet away from anything that can burn (75%). Following the Remembering WhenTM program, more participants understood how to use stop, drop and roll procedures (baseline: n=72, 66%; follow-up: n=93, 85%), and installed smoke alarms on every level of the home (baseline: n=86, 78%; follow-up: n=107, 97%) and inside each bedroom (baseline: n=83, 76%; follow-up: n=95, 86%). We did not observe differences in the percentage of participants improving their fire safety practices between the two study arms.

• Perceptions about falls and fire prevention: Perceptions about the participants’ susceptibility to experience a house fire increased slightly, whereas perceptions about the severity of experiencing a house fire and fear decreased after the program. Participants’ perceptions about having control over their fire prevention behaviors improved after program. They generally reported very low levels of perceived susceptibility to experiencing a fall, and felt that the outcome of a fall can be somewhat serious. These perceptions did not change after program participation. However, participants reported improved perceptions about their own ability to prevent falls, to engage in falls prevention behaviors, and felt they had control over their preventive behaviors after the program participation. At the end of the project, participants also reported an increase in the perceived availability of informational support about falls and fire prevention.

In addition to the aims of the study, we also conducted qualitative interviews with the older adult participants and Outreach Team members about their satisfaction with the program and suggestions for improvement. These findings are provided in the final report.

—— Page 4 ——

Project Background Falls are the leading cause of unintentional fatal and nonfatal injury among older adults in the United States (National Center for Injury Prevention and Control, 2013). More than one-third of older adults fall each year, and many of these falls result in decreased or permanent loss of functioning, loss of independence and a fear of falling (Centers for Disease Control and Prevention, 2008). Older adults also have one of the highest rates of death from house fires, compared to their younger counterparts (Flynn, 2010). Older adults are more likely to suffer from reduced sensory abilities and mental health, which reduces reaction time and puts them at higher risk for causing a fire and dying (US Fire Administration, 2006). In Iowa, rates of fall-related deaths among older adults far exceed the national average, while fire-related death rates are comparable to the national average (National Center for Injury Prevention and Control, 2013).

Numerous epidemiologic and efficacy studies have demonstrated that falls and fire risks among older adults can be reduced (Robertson & Gillespie, 2013; Marshall et al., 1998). The literature on fall risk reduction interventions and behaviors is extensive, whereas the literature on fire prevention, particularly among older adults, is more limited. In one meta-analysis study of 159 randomized controlled trials (RTCs) of falls prevention interventions for older, community-dwelling adults, exercise programs were found to be consistently effective in reducing the risk of falls (Robertson & Gillespie, 2013). Home safety programs, particularly those delivered by occupational therapists, were effective primarily for individuals already at higher risk of falling. Similarly, vitamin D supplementation benefitted only those who were deficient. There was also evidence to support gradual withdrawal of psychotropic medications and education for medical professionals on improved prescribing. Overall, the effectiveness of using educational materials (such as health risk appraisal using tailored feedback) or cognitive behavioral interventions (e.g., fall risk assessment combined with an exercise program) for falls prevention could not be established.

A subsequent umbrella review of 16 falls-prevention meta-analyses, including that described by Robertson and Gillespie (2013), confirmed, but also qualified, some of the meta-analysis findings (Stubbs et al., 2015). Again, the authors found consistent evidence that exercise interventions are effective in reducing falls among older adults. They also found that interventions that are tailored to the individual older adult and include multiple components (e.g., requiring a multidisciplinary team, a variety of assessments, and referral procedures) are also consistently effective. On the other hand, this umbrella review uncovered conflicting evidence for the benefit of environmental interventions, but found that home assessment and modification interventions delivered by an occupational therapist are effective. In sum, exercise-based interventions and those which are personalized appear to be the most beneficial for older adult falls prevention.

National fire statistics point to cooking, careless misuse of materials or heat sources, and “open flames” as the top three causes of residential building fires resulting in injuries between 2011 and 2013 (US Fire Administration, 2015). While these causes point to behavioral intervention targets, to our knowledge, there have been no meta-analyses of fire prevention interventions or protective behaviors to date; however, some published observational and quasi-experimental studies point to the importance of

—— Page 5 ——

smoke alarm-related behavior. In a retrospective analysis of fire deaths in North Carolina, Marshall and colleagues (1998) determined that having a smoke detector and having a potential rescuer in the home during a “multiple persons home” fire were protective against dying in a fatal fire. The “rescuer” factor was particularly important for individuals who are vulnerable due to age or physical or cognitive impairment. Some fire prevention behavioral interventions have incorporated smoke alarm behaviors (e.g., smoke alarm placement, appropriate response to an alarm, and maintenance) and services. One quasi-experimental study evaluated a firefighter-delivered, door-to-door educational intervention, in which firefighters spoke with residents about fires, gave them written information, offered to install a smoke alarm for free, and informed residents that they could request a free home safety inspection (Clare et al., 2012). Relative to the comparison group, the group receiving the intervention showed a sharper decrease in the rate of fires and sharper increases in smoke alarm activation during fires and fires being contained in the room of origin. Finally, a review of US fire prevention interventions demonstrated that, although the quality of the original studies reviewed was low-to-moderate, there is evidence to suggest that smoke alarm distribution programs can lead to a reduction in fire injuries relative to comparison groups (Ta et al., 2006). Furthermore, the authors noted that using a canvassing method of smoke alarm distribution was the most effective method for reaching target households.

Remembering WhenTM, a program developed by the National Fire Protection Association (NFPA) and CDC, has strong capacity for program dissemination and implementation and is well-recognized as a community-based model of falls and fire prevention for older adults. The program has 16 key falls and fire prevention messages, the majority of which target residential safety hazards and behaviors (Remembering When, http://www.nfpa.org). The Remembering WhenTM program is delivered in both group and home visit formats by teams of fire department officials and home visiting organizations. The program aims to facilitate preventive behaviors through highlighting the perceptions of risks associated with falls and fire and providing recommendations on ways to reduce such risks. The Remembering WhenTM program was evaluated about 10 years ago for changes in knowledge about preventive behaviors (e.g., how to stop a grease fire, how to reduce the risk of falling in the bathtub or shower), comparing older adults who received the program to those who did not (Ottoson et al., 2004). Five months post-program delivery, older adults experienced a significant increase in knowledge of both fall and fire prevention, compared to older adults who did not receive the program. This evaluation, however, did not evaluate the changes in falls prevention and fire safety risks and behaviors. The evaluation also under-represented lower income and homebound older adults, who are at greater risk for falls and residential fires (Jones et al., 2014; Herr et al., 2013).

The objective of the current pilot evaluation study was to reduce the risk of falls and house fires among community-dwelling older adults in five communities in Iowa. The specific aims were to evaluate the effectiveness of the Remembering WhenTM program, comparing program delivery formats and also comparing homebound with non-homebound older adults, in: (1) Improving perceptions associated with falls and house fires, and (2) Changing falls prevention and fire safety behaviors. Furthermore, this pilot evaluation study was intended to provide lessons for future, large-scale evaluations of the Remembering WhenTM program, as well as identify how the program could be potentially improved.

—— Page 6 ——

Methodology

1. Overview The Remembering When™ Pilot Evaluation Study was conducted between February and November, 2015 in five communities across Iowa. The University of Iowa Institutional Review Board approved the study. The five study sites were selected based on local fire department completion of Remembering When™ training and interest in participating in the study. Within each community, the research team collaborated with up to two fire department officials comprising an Outreach Team, which was primarily responsible for delivering the Remembering When™ program, providing data on delivery activities and providing feedback in a subsequent exit interview. In addition, adults ages 65 and older were recruited in each community to participate in the study by the study staff and participated in a series of telephone interviews conducted before and after participation in Remembering When™ training. Study data were entered and managed using REDCap electronic data capture tools hosted at the University of Iowa (Harris et al., 2009). REDCap (Research Electronic Data Capture) is a secure, web-based application designed to support data capture for research studies, providing 1) an intuitive interface for validated data entry; 2) audit trails for tracking data manipulation and export procedures; 3) automated export procedures for seamless data downloads to common statistical packages; and 4) procedures for importing data from external sources.

2. Outreach Teams

2.a. Recruitment of Outreach Teams Prior to the start of the study, a representative of the National Fire Protection Association invited fire departments in Iowa communities to send at least two representatives to an official Remembering When™ training held in Des Moines, IA, in March of 2015. Representatives from eight Iowa fire departments attended the training and were invited to participate in the study. Eligibility criteria for participation as an Outreach Team member in the study included being a member of a community organization that serves older adults, having been trained in the Remembering When™ program, and being based in the state of Iowa. Ten individuals from five Iowa fire departments gave verbal consent over the telephone to participate in the study. Most had been working with community service programs and with older adult populations for more than 10 years. Only one individual had experience delivering the Remembering When™ program prior to the study. Each Outreach Team member was offered $250 in two installments (two Outreach Team members declined due to departmental policy) for completing all study components, an additional $50 from the NFPA’s Public Education Division, and an invitation from the Project Manager of the Public Education Division to attend the 2015 NFPA Conference and Expo in Chicago with conference registration paid.

2.b. Remembering When™ Program Delivery Outreach Team members were primarily responsible for ensuring that older adult study participants received Remembering When™ training in home visit and in group presentation format for select participants. Outreach Teams were given older adult participant contact information to schedule Remembering When™ home visit trainings directly with the participants in their home, whereas the research team coordinated scheduling of the group presentations. Outreach Team members completed

—— Page 7 ——

a program delivery log after each home visit or group presentation, recording the degree to which they covered each of the key Remembering When™ messages, the degree of attentiveness of the participant(s), any issues encountered and their resolutions, the length of the training, and any materials or activities used. To assess how well Outreach Team members completed the program delivery logs, members of the University of Iowa (UI) research team observed two home visits and at least one group presentation by each Outreach Team. The research team members recorded their observations on an observation log, which mostly mirrored the program delivery log content. The Outreach Teams delivered the program messages as designed but had varying degrees of program materials and props that were used during the presentations (see Section 1 in Results).

Out of the five Outreach Teams that enrolled, four completed all program delivery events. One team was unable to continue due to staffing constraints but did co-facilitate two home visits and one group presentation in the community with support of another outreach team in a nearby community.

2.c. Exit Interview Once the active study period for Outreach Teams concluded, those who had directly participated in program delivery were invited to participate in a final exit interview, either individually or as a team. The exit interview guide included questions about participating in the study project overall; participating in the initial Remembering When™ training; their experiences delivering the program; implementing the study protocols and data collection instruments; communicating with the research team; adequacy of their incentives; and suggestions for improving their experience in the study and delivering the program. Because questions included feedback about the research team, an independent interviewer was hired to conduct the final interview. In total, six exit interviews with seven Outreach Team members from all five Outreach Teams were audio-recorded and transcribed.

3. Older Adult Participants

3.a. Recruitment and Randomization In order to identify appropriate locations and methods for recruiting older adult participants in each community, the Outreach Team members were asked to recommend community organizations for the research team to approach. Overall, 44 community organizations, identified by Outreach Team recommendations and web searching, were approached to collaborate in recruiting older adults to participate in the study. Active recruitment of older adults occurred in 13 community organization sites, and another six organizations facilitated passive recruitment using brochure or email announcement distribution. Organizations approached include senior residential complexes (not including assisted living or nursing home complexes), home visiting and meal delivery organizations, congregate meal sites, and places of worship (Figure 1). In addition to active and passive recruitment via community organizations serving older adults, one Outreach Team posted a social media announcement about the study, two Outreach Teams distributed brochures to older adults at community events, and one community periodical published a study announcement to its subscribers. The majority of participants who completed the study were recruited at an in-person recruitment event (n=79), through home mailings (n=14), or via email or magazine announcements (n=11).

—— Page 8 ——

Figure 1. Older adult recruitment by community organization type.

Older adults were recruited by the research team at the organizations agreeing to collaborate. The research team, oftentimes with members of the Outreach Teams, made a presentation about the study and program at organization events or set up a table in high-traffic areas of the organization. Older adults interested in participating were screened into the study based upon the following eligibility criteria: at least 65 years of age or older; not wheelchair-bound or bed-bound; able to hear well enough to participate in the Remembering When™ program and telephone interviews; cognitively able to participate in research, as measured by the six-item Mini Mental State Examination (Callahan et al., 2002); has a working telephone number where he/she can be reached; speaks and understands English; resides within a reasonable driving distance of the Outreach Team’s jurisdiction, as determined by the Outreach Team members; lives independently (i.e., not in assisted living or a nursing home); and has not participated in Remembering When™ training before. Active and passive recruitment was conducted in collaboration with a variety of community organizations (Figure 1) in each of the five study communities. Older adult participants enrolled in the study by giving verbal informed consent either in person at a recruitment event or over the telephone.

Overall, 184 older adults were screened for eligibility, and 151 enrolled in the study (Figure 2). Each enrolled participant was assessed for homebound status by asking how often they left their home to go outside in the past month. If participants left their homes rarely (one day or less per month), they were

0

2

4

6

8

10

12

14

16

18N

umbe

r of O

rgan

izat

ions

Community Organization Type

Decline/Unable to Contact

Passive Recruitment

Active Recruitment

—— Page 9 ——

considered homebound (Ornstein et al., 2015). Six participants met homebound status, while the remaining 145 were defined as non-homebound. Non-homebound participants (n=145) were randomly assigned to one of two study arms: (1) individual receiving a Remembering When™ home visit (n=70), and (2) individual receiving both the Remembering When™ group presentation and home visit (n=75). The six enrolled participants who met homebound status were allocated to the home visit study arm. However, one of the six attended a group presentation and was therefore re-allocated to the group presentation and home visit study arm.

Of the 145 non-homebound enrolled participants, 70 were randomized to study arm one and 75 were randomized to study arm two (Figure 2). Couples who were members of the same household were eligible to enroll (n=22 individuals) and were assigned to study arms based on the results of randomization for one member of the couple. Ten participants were subsequently withdrawn as “ineligible” because they were unable to attend a scheduled Remembering When™ group presentation to which they were assigned. Subsequently, the protocol was changed to retain participants who had missed the group presentation (n=9) and instead treat them as home visit-only participants. Six participants withdrew from the study, mostly citing unspecified personal reasons, with one participant expressing wariness about having a home visit. One participant died before participating in the follow-up interview. We broke the randomization scheme to retain study participants: (1) who were assigned to study arm one but attended a Remembering When™ group presentation (n=7), and (2) who did not attend a group presentation but were assigned to study arm two (n=9). The former participants were re-assigned to study arm two, and the latter were re-assigned to study arm one. In the final analysis, only one individual from each enrolled cohabitating couple (n=11) was randomly selected to have their data included in the final dataset (n=110) in order to eliminate bias from dependency within couples. All six of the homebound participants were retained for analysis (i.e., none were lost to follow-up and none were from cohabitating couples).

3.b. Data Collection Each enrolled older adult participated in three structured, audio-recorded telephone interviews (Figure 3): Baseline Interview(s), Post-Program Satisfaction Interview, and Follow-Up Interview. Each interview ranged in length from approximately twenty to forty-five minutes. Each older adult participant received $75 in compensation, issued in three installments, one after completing each interview (except the second baseline interview, see below).

Baseline Interview(s) The first forty participants participated in two baseline telephone interviews, the first one within one to two weeks of enrollment in the study and the second approximately one month following the first. Subsequent participants completed only one baseline interview. We incorporated a second baseline to assess whether the interview itself may be influencing changes in fall and fire behaviors recommended in the Remembering When™ program. A comparison between baseline 1 and 2 responses of selected variables was conducted with the initial forty interview responses. Weighted Kappa coefficients were calculated for variables measured on a Likert scale, and simple Kappa coefficients were calculated for variables measured as binary. The variables chosen were related to falls and fire behaviors, self-efficacy and social support. Kappa coefficients less than 0.41 were considered indicators of only fair agreement

—— Page 10 ——

between the first and second baseline interviews, suggesting that there may be changes in participant responses following the first baseline interview (van Belle et al., 2004). The calculation of Kappa coefficients for some variables was somewhat limited due to low variation in response distributions. Taking this into consideration, most Kappa coefficients suggested that participants did not change their behaviors. However, there was some indication that they may be slightly more confident in engaging in safety precautions and that they engaged in more social support after the first baseline interview. Taken together, with reported interviewee fatigue of participating in an additional interview, the second baseline interview was eliminated for the remaining participants. A total of 149 older adults completed the first baseline interview.

Figure 2: CONSORT enrollment and retention flow chart.

Upon completion of the baseline interview(s), every older adult participant was contacted by the Outreach Team in their community to schedule a Remembering When™ home visit training and/or

—— Page 11 ——

group presentation. All except six of the study arm two participants attended a Remembering When™ group presentation training prior to the home visit. A total of 133 older adults received training in the Remembering When™ program.

Figure 3. Timeline of older adult participant interviews and Remembering When™ training.

Post-Program Satisfaction Interview Within approximately one to two weeks (average 5 days, range = 0 – 28 days) of completing all Remembering When™ training components, each enrolled older adult participated in a Post-Program Satisfaction telephone interview. Among the 133 participants receiving program training, 132 (99%) completed the post-program satisfaction interview.

Follow-Up Interview Approximately one to two months after the post-program interview, 130 participants completed a Follow-Up telephone interview. The average number of days between receiving the Remembering When™ program and the follow-up interview was 50 days (range= 12 – 106 days).

3.c. Measurements

Baseline and Follow-Up Interviews The baseline and follow-up interviews collected data on fall and fire risks, falls prevention and fire safety behaviors and perceptions regarding falls and fire risks and preventive behaviors.

Fall and Fire Risks Demographics of the participants were collected, including age, gender, race and ethnicity.

Risk factors specific to falls. Validated measures of falls risk and fear of falling were collected. The Self-Rated Fall Risk Questionnaire for Older Adults was used to measure whether participants were at risk for falling (Rubenstein et al., 2011). The questionnaire follows the American Geriatrics Society / British Geriatrics Society guidelines to assess independent predictors for falls (AGS/BGS, 2011). Participants were asked to respond to ‘yes/no’ questions about their: history of falls in the past six months, fear of falling, urinary incontinence, depression, current medications, use of an assistive device, balance and gait, lower extremity muscle strength, loss of sensation in extremities, and visual acuity. The Self-Rated Fall Risk Questionnaire is scored by summing all ‘yes’ responses. A score of at least 4 (range 0-12) suggests that the older adult is at risk for a fall.

The Falls Efficacy Scale was used to measure fear of falling (Tinetti et al., 1990). Participants were asked to rate their confidence in performing activities without falling, on a scale from 1=very confident to 10=not at all confident. The items are summed to create a total score, where a score greater than 70

Baseline 1 Interview (n=149)

Baseline 2 Interview

(n=40)

Remembering When Training

(n=133)

Post-Program Satisfaction Interview (n=132)

Follow-Up Interview (n=130)

—— Page 12 ——

suggests a fear of falling (Tinetti et al., 1990). Use of a cane or walker (always, sometimes, never) was also collected.

Risk factors specific to fires. Data on risk factors for fires was collected and included annual household income (< $50,000, $50,000+), home structure (house, apartment/condominium, modular/manufactured, duplex/townhome), residential tenure (owner-occupied, rental), current smoking status, and alcohol consumption in the last 30 days (Marshall et al., 1998; Runyan et al., 2004)

Falls Prevention and Fire Safety Behaviors Falls prevention and fire safety behaviors were collected using the “Home Safety Checklist” provided in the Remembering WhenTM program training manual. Response options for these questions, however, were modified from a ‘yes/no’ format to capture the extent or frequency with which participants engaged in falls and fire safety behaviors. For example, a five-point Likert scale ranging from “to a very great extent” to “to a very small extent” was used to rate the extent to which participants engaged in such behaviors as keeping walking areas and stairways free of tripping hazards and inspecting electrical cords for damage. Five-point Likert scales ranging from “all of the time” to “never” were used to measure the frequency with which participants engaged in behaviors such as taking time getting up, staying in the kitchen when cooking and using grab bars in bathrooms.

Responses to each of the behaviors in the Remembering WhenTM program were categorized into “Meets Recommendation” and “Does Not Meet Recommendation”, as defined below:

Falls Prevention Behavior

Definition for Meeting Remembering When Recommendation

Exercise Regularly Exercise at least 3 times a week Take Your Time Take all the time you need to get up from sitting or lying down without rushing

Question: How often do you take all the time you need to get up from sitting or lying down? Response: “All of the time” or “Most of the time”

Keep Stairs and Walking Areas Clear Keep stairs and walking areas free of electrical cords, shoes, clothing, books, magazines and other treasures

Question: To what extent are the stairs and walkways in your home covered by paper, books, or other household items? Response: “To a very small extent” or “Not at all”

Improve the Lighting in and Outside Your Home Use nightlights or a flashlight to light the path between your bedroom and bathroom

Question: To what extent is the path between your bedroom and bathroom well lit at night? Response: “To a very great extent” or “To a great extent”

Use Non-Slip Mats Use non-slip mats in the bathtub and on shower floors Have grab bars installed on the wall next to the bathtub, shower and toilet

Question: Do you use non-slip mats or other non-slip surfaces in the bathtub or shower that you use the most? Response: “Yes” Question: Do you use non-slip mats or other non-slip surfaces in the bathtub or shower that you use the most? Response: “Yes”

Be Aware of Uneven Surfaces Use only throw rugs that have rubber, non-skid rug pads

Question: How many throw rugs in your home have a rubber, non-skid backing? Response: “All of them”

Stairways Should be Well Lit

—— Page 13 ——

Turn on the lights before using the stairs Have sturdy handrails on both sides of the stairs

Question: How often do you turn on the lights before you use the stairs? Response: “All of the time” Question: Do the stairs in your home have sturdy handrails? Response: “Both side of stairs”

Wear Sturdy, Well-Fitting Shoes Wear sturdy, well-fitting shoes with non-slip soles

Question: Which of the following best describes the kind of shoes that you wear most regularly in the home? Response: “Low-heeled shoes”

Fire Safety Behavior

Definition for Meeting Remembering When Recommendation

Give Space Heaters Space Keep space heaters at least 3 feet away from anything that can burn- including you Turn off space heaters when you leave your home or go to bed

Question: Do you keep space heaters at least 3ft away from all people, furniture, or objects? Response: “Yes” Question: How often do you shut off the space heater when you leave the room? Response: “All of the time”

Stay in the Kitchen When Frying Food Wear tight-fitting, rolled up, or short sleeves when cooking Stay in the kitchen when you fry

Question: When you are cooking food, which of the following best describes what you are usually wearing? Response: “Tight long-sleeves” or “Tight short-sleeves” Question: How often do you stay in the kitchen when you fry? Response: “All of the time” or “Most of the time”

If Your Clothes Catch Fire: Stop, Drop and Roll

Question: What would you do if your clothes were to catch on fire? Response: “Stop, drop, and roll”

Smoke Alarms Save Lives Have smoke alarms installed on every level of your home Have smoke alarms installed inside each bedroom Have smoke alarms outside each sleeping area Test alarms monthly

Question: Do you have a smoke alarm at each level of your home? Response: “Yes” Question: Do you have a smoke outside each bedroom in your home? Response: “Yes” Question: Do you have a smoke alarm in each bedroom of your home? Response: “Yes” Question: How often do you or someone test the smoke alarms in your home? Response: “About once a month”

—— Page 14 ——

Plan Your Escape Around Your Abilities Have a landline telephone or cell phone and charger near your bed Consider subscribing to a medical alert system

Question: Do you have a telephone in your bedroom? Response: “Yes” Question: Do you have a medical alert button? Response: “Yes”

In addition to categorizing each of the Remembering WhenTM program behaviors into whether or not the participant met the recommendation as specified in the program, the behaviors were also categorized into whether they “Improved”, “Stayed the same” or “Declined” between the baseline and follow-up periods.

Perceptions Perceptions regarding falls and related preventive behaviors were assessed using concepts of the Health Belief Model (HBM). The HBM has been extensively used to explain why individuals take actions to prevent injuries or illnesses (Champion and Skinner, 2008), and has been used in developing interventions for falls prevention and home fire safety (Hill et al., 2009). In the context of falls prevention, the HBM posits that individuals are likely to undertake falls prevention behaviors if: (1) they are aware of their own risks (perceived susceptibility and seriousness), (2) if the recommended actions are perceived to be within their ability to implement them (high self-efficacy) and these actions are seen as being of benefit to them (high perceived benefits), (3) they have the resources to take the actions (barriers addressed), and (4) the perception of benefits outweigh the costs (WHO, 2007).

Perceived susceptibility was assessed by asking participants to rate their perceived likelihood of experiencing a fall/house fire in the next 6 months (5-point scale ranging from “not at all likely” to “extremely likely”), and perceived severity was assessed by asking how serious it would be if (he/she) fell/had a house fire (5-point scale ranging from “not at all serious” to “extremely serious”). The overall measures of self-efficacy asked participants how sure they were that they could keep themselves from experiencing a fall/house fire (“not at all sure” to “extremely sure”). The overall self-efficacy regarding the behaviors were measured by asking participants to rate the confidence that they could engage in the safety precautions that may help reduce falls/fire risks. Perceived benefits were assessed by asking how likely taking precautions or making changes to the home environment would help reduce their chances of experiencing a fall/fire. Perceived barriers were assessed by asking the extent to which participants agree or disagree with statements that there are things and situations beyond their control that make it difficult to take safety precautions to reduce falls/fire risks (“not at all” to “to a very great extent”). In addition, participants were asked to rate the extent to which social support is available to help them engage in falls prevention and fire safety behaviors using a 5-point scale (“not at all” to “to a very great extent”). Whether participants talk about falls/fire prevention with family and friends, health care professionals, and senior center staff (yes/no) was also assessed at baseline and follow-up.

Post-Program Satisfaction Interview The post-program satisfaction interview included items regarding their satisfaction of the Remembering When™ training (e.g., convenience of group and home visit formats, length) and program (e.g.,

—— Page 15 ——

relevance, quality, understandability, knowledge gained); feedback on the Outreach Team member presenting the material (e.g., knowledge, helpfulness, friendliness, trusted source); feedback on what they felt was most useful, least useful and how they might change the program and/or training (e.g., suggestions for improvement, alternative modes of program delivery, alternative presenters), and; perceptions and self-efficacy questions as described in the previous “Perceptions” section.

3.d. Data analysis

Older adult interviews Fall Prevention and Fire Safety Behaviors Chi-square tests were used to examine the association between categorical fall and fire risk factors and the study arms (home visit only, home visit and group presentation); Fisher’s exact test was used to examine these associations when cell counts dropped below 5. Independent group t-tests were used to examine the association between continuous fall and fire risk factors and the study arms. McNemar’s exact test was used to compare the proportion of participants engaged in the Remembering WhenTM falls and fire safety behaviors between the baseline and follow-up periods. Chi-square tests (or Fisher’s exact tests for small cell counts) were used to compare the proportion of participants who improved their behaviors between baseline and follow-up across the study arms.

Perceptions about Falls/Fire, Associated Preventive Behaviors and Social Support To assess the levels of perceptions about falls/fire and associated preventive behaviors, frequencies for each of the response options were first examined. For the items that resembled normal distributions of the responses, descriptive statistics (i.e., means, standard deviations, ranges) were examined and presented for the entire sample (n=110) as well as for each study arm: home visit only (n=64) and home visit plus group presentation (n=46). Potential changes in levels of reported perceptions (i.e., susceptibility, severity, fear, self-efficacy, benefits, intention, control/barriers) as well as perceived available social support (i.e., to reduce fall/fire, informational, emotional, tangible) between baseline and follow-up were assessed by conducting paired sample t-tests for each construct. These analyses were conducted for the entire sample (n=110) as well as for each group separately. Differences in the extent of changes in these perceptions between two groups were evaluated by conducting independent sample t-tests to compare change scores (follow-up minus baseline).

Discussions about Falls/Fire Prevention with Others For the discussion of falls/fire prevention with others, the number of participants reporting to have discussed with family and friends, health care professionals, and senior center staff were compared at baseline and follow-up using McNemar’s exact test.

Post-follow-up interview comments Audio-recorded comments made by some participants after the follow-up interview regarding their participation in the study and Remembering When™ were transcribed and categorized by theme, such as “General appreciation,” “Learned something new,” and “Suggestions for improvement.” Out of 130 final interviews, 115 audio files were available for thematic analysis. Instances of the themes were tallied to yield counts of occurrence.

—— Page 16 ——

Outreach team exit interviews Outreach Team exit interviews were transcribed non-verbatim, with false starts and filler words omitted, by a transcription service (Landmark Associates, Inc.). All transcripts were imported and analyzed using Atlas.ti qualitative analysis software version 7.5.9. One member of the research team used an iterative process of transcript reading, coding, and memoing to analyze the interviews. The initial codebook was comprised of topical codes based on the exit interview guide. Example codes include “Group presentations,” “Home visits,” “Most effective program delivery strategy,” and “Data collection.” After these topical codes were applied to the transcripts, the researcher wrote memos and developed additional codes based on emergent themes such as, “Alignment of Remembering WhenTM with fire department goals,” “Outreach Team time and staffing,” and “Quality of interactions with older adults.” Through this iterative process, a series of emergent themes were explored using memos, tables, and other analytic products.

Results

1. Remembering When™ Program Delivery Using 124 home visit program delivery logs and nine group presentation program delivery logs completed by Outreach Team members, the Outreach Teams appeared overall to deliver the Remembering When™ key messages with fidelity. There was variation in the extent to which Outreach Team members covered the key fall and fire safety messages and used supportive materials, but some of this variation appears to have been due to the different needs of the participants, as the Outreach Teams reported tailoring the program contents as needed in their exit interviews. For instance, information about safe smoking practices was the least covered of all fire and falls messages, most likely because only four older adult participants were smokers. In the group presentations, Outreach Teams reported covering all key messages to “some extent” or “to a great extent” most of the time. On average, group presentations lasted 56 minutes and included nine attendees; home visits lasted 49 minutes on average. There was wide variation in the use of props for group presentations (e.g., space heater, grab bar, fire alarm), but most Outreach Teams used Remembering When™ program materials such as message cards and the home safety checklist in both group and home presentations. Most reports indicated that participants were very or completely attentive during group presentations and home visits. Data from thirteen observation logs completed by the research team were largely consistent with the program delivery log data provided by the Outreach Teams; slight differences in the extent-of-coverage ratings for each topic likely were due to different interpretations of the rating scale.

2. Older Adult Study Population

2.a. Demographics, Housing and Health Characteristics of Enrolled Study Population (n=130) More than half of the participants (66%) were at least 75 years of age, and the majority were female (76%), White (98%) and not of Hispanic or Latino origin (99%) (Table 1). Among those who reported a household income, more than two-thirds were below a $50,000 annual income. Sixty percent of the participants lived in an apartment or condominium, and approximately one-third (34%) lived in a house.

—— Page 17 ——

There was a fairly even distribution of participants who owned their home (52%) and those who rented (48%).

Most of the participants were non-smokers and/or lived in a household without a smoker (95%), and about two thirds (63%) reported not having consumed an alcoholic beverage such as beer, wine, malt beverage or liquor in the past 30 days (Table 1). Overall, just over half of the participants (55%) reported being in very good or excellent health, while another 35% reported being in good health.

2.b. Demographics and Fall and Fire Risks at Baseline among Sample with Randomly Selected Cohabitating Individual (n=110), Overall and by Study Arm Older adult participants exhibited risk factors for falls, including older age, female gender and overall multifactorial risk based on a validated fall risk self-assessment (Table 2). On average, participants were 77.6 years of age (sd= 7.62), and 81% were female. The fall risk self-assessment was used to calculate a fall risk score, where a score of 4 or more (range= 0-12) indicates that an older adult may be at risk for a fall (Rubenstein et al., 2011). In our study population, just under half of the participants (48%) had a risk score of 4 or more. In contrast, two-thirds of the participants never used a cane or walker. In addition, participants in both study arms had low fear of falling (mean scores (sd): home visit= 16.7 (10.0); home visit and group presentation= 14.9 (7.78)) based on the Falls Efficacy Scale, where a total score exceeding 70 indicates a fear of falling (Tinetti et al., 1990). Older adult participants also exhibited risk factors for experiencing a house fire, including older age, lower household income, and rental residential tenure (Table 2). Nearly 70% of the participants had less than a $50,000 per year income, and about half (51%) lived in a rental property. However, very few participants smoked or lived with someone who smoked (6%). We found no differences in fall and fire risks between participants in the home visit only study arm and participants in the home visit and group presentation study arm (p > 0.05) (Table 2).

Ten percent of the participants reported ever having participated in a falls prevention program, and 18% reported ever having participated in a fire safety program (Table 2). None of the participants had received the Remembering WhenTM program prior to study enrollment (per study eligibility criteria).

—— Page 18 ——

Table 1. Demographic, housing and health characteristics, all participants, n=130 Demographic Characteristic N (%) Age 65-74 75-84 85+

44 (33.8) 58 (44.6) 28 (21.5)

Gender Male Female

32 (23.8) 99 (76.2)

Race White Non-White / Multiracial

127 (97.7)

3 (2.3) Ethnicity Hispanic/Latino Not Hispanic/Latino Missing

0

129 (99.2) 1

Household income < $50,000 per year $50,000+ per year Unknown Refused

87 (66.9) 32 (24.6)

4 (3.1) 7 (5.4)

Structure of home House Apartment / Condominium Modular / Manufactured Housing / Mobile Home Duplex / Townhome

44 (33.8) 78 (60.0)

3 (2.3) 5 (3.8)

Residential tenure Owner-occupied Rental Unknown

67 (51.5) 62 (47.7)

1 (0.8) Smoker or Someone in home smokes Yes No

6 (4.6)

124 (95.4) Consumed alcohol in last 30 days Yes No Unknown

47 (36.2) 82 (63.1)

1 (0.8) General health Excellent Very Good Good Fair Poor

17 (13.1) 54 (41.5) 45 (34.6)

2 (1.5) 12 (9.2)

—— Page 19 ——

Table 2. Frequency of demographics and fall and fire risk factors at baseline, n=110

Total

Home Visit Only Home Visit and

Group Presentation

p-value1 Age, mean (sd) 77.6 (7.62) 77.8 (7.46) 77.3 (7.90) 0.7262 Gender, n(%) Male Female

21 (19.1) 89 (80.9)

9 (14.1)

55 (85.9)

12 (26.1) 34 (73.9)

0.114

Race, n (%) White Non-White

108 (98.2)

2 (1.8)

62 (96.9)

2 (3.1)

46 (100.0)

0

Household income, n (%) < $50,000 per year $50,000+ per year Unknown Refused

76 (69.1) 25 (22.7)

2 (1.8) 7 (6.4)

50 (78.1) 12 (18.8)

1 (1.6) 1 (1.6)

26 (56.5) 13 (28.3)

1 (2.2) 6 (13.0)

0.1133

Structure of home, n (%) Modular Not Modular

33 (30.0) 77 (70.0)

18 (28.1) 46 (71.9)

15 (32.6) 31 (67.4)

0.613

Residential tenure, n (%) Rent Own Unknown

56 (50.9) 53 (48.2)

1 (0.9)

34 (53.1) 30 (46.9)

22 (47.8) 23 (50.0)

1 (2.2)

0.7363

Smoker or Someone in home smokes, n (%) Yes No

6 (5.5) 104 (94.5)

4 (6.3) 60 (93.7)

2 (4.3) 44 (95.6)

0.5064

Consumed alcohol in last 30 days, n (%) Yes No

41 (37.3) 69 (62.7)

20 (31.3) 44 (68.8)

21 (45.7) 25 (54.3)

0.162 4

General health, n (%) Excellent/Very Good Good/Fair/Poor

62 (56.4) 48 (43.6)

35 (54.7) 29 (45.3)

27 (58.7) 19 (41.3)

0.701 4

Use cane or walker, n (%) Always Sometimes Never

17 (15.4) 21 (19.1) 72 (65.5)

8 (12.5)

15 (23.4) 41 (64.1)

9 (19.6) 6 (13.0)

31 (67.4)

0.298

Fall risk5, n (%) At risk for a fall Not at risk for a fall

53 (48.2) 57 (51.8)

33 (51.6) 31 (48.4)

20 (43.5) 26 (56.5)

0.403

Fear of falling, mean (sd) 15.9 (4.14) 16.7 (10.00) 14.9 (7.78) 0.5392 Participation in falls prevention program, n (%) Yes No

11 (10.0) 99 (90.0)

7 (10.9) 57 (89.1)

4 (8.7) 42 (91.3)

0.7584

—— Page 20 ——

Participation in fire safety program, n (%) Yes No

20 (18.2) 90 (81.8)

14 (21.9) 50 (78.1)

6 (13.0) 40 (87.0)

0.236

1: Chi-square test, comparing study arms; 2: Independent group t-test, comparing study arms; 3: Does not include Unknown and/or Refused responses; 4: Fisher’s exact test, comparing study arms; 5: Based on Fall Risk Self-Assessment Tool (Rubenstein et al., 2011)

2.c. Fall and Fire Prevention Behaviors among Sample with Randomly Selected Cohabitating Individual (n=110) Before asking about specific behaviors related to the Remembering WhenTM program at baseline, participants were asked to describe the types of activities they currently engaged in to reduce their risk of falling and experiencing a home fire (Table 3). For falls prevention practices, the most common behaviors mentioned were “being careful” (41%), followed by keeping walking paths clear from potential trip hazards (34%), removing/not owning throw rugs or using a rubber backing (22%), and using a walker or cane when needed (21%). For fire safety practices, the most frequent behaviors mentioned were having, installing and/or checking smoke alarms (28%) and “being careful” (27%). About 20% of the participants also mentioned not burning candles unattended or rarely using them (22%), and turning off burners and the oven when not in use (21%). Overall, only a small percentage of the participants offered behaviors that are recommended in the Remembering WhenTM program.

Table 3. Frequency of reported baseline falls prevention and fire safety practices, n=1101

Current Falls Prevention Behavior

Frequency of Mention

N (%) Current Fire Prevention Behavior

Frequency of Mention

N (%)

Being careful 45 (40.9) Having, installing, and/or checking smoke alarms 31 (28.2)

Keep walking paths clear, removing potential trip hazards 37 (33.6) Being careful 30 (27.3) Removing/not owning throw rugs, using rubber backing 24 (21.8)

Not burning candles unattended or rarely using them 24 (21.8)

Using a walker or cane as needed 23 (20.9) Turning off burners and the oven when not in use 23 (20.9)

Using or installing handrails 20 (18.2) Staying in the kitchen most/all of the time when cooking 18 (16.4)

Exercising 19 (17.3) Keeping objects away from stove 16 (14.5) Leaning on furniture or other objects for support 12 (10.9)

Unplugging or turning off appliances when leaving home 16 (14.5)

Using grab bars 10 (9.1) Not cooking or using the stove much 15 (13.3)

Being aware of the possibility of falling 9 (8.2) Not smoking 12 (10.9) Taking time getting up and moving 8 (7.3)

Checking or reducing overload of electrical plugs and cords 12 (10.9)

—— Page 21 ——

Being mindful of pet hazards (tripping on toys, pets themselves) 6 (5.5)

Removing or minimizing flammable items and clutter 11 (10.0)

Wearing shoes in the house 5 (4.5) Having a fire extinguisher 10 (9.1) Keep pathways lit in house 5 (4.5) Having an electric stove 9 (8.2)

Using a shower seat or bench 5 (4.5) Not or rarely using extension cords 6 (5.5)

Using caution when stepping on objects for height (e.g., ladders) 4 (3.6) None/don't know 6 (5.5) Nothing/don't know what to do 3 (2.7) Using a timer when cooking 3 (2.7) Using a non-slip mat or surface in the shower or bathtub 2 (1.8)

Being careful using fireplace or not using it at all 3 (2.7)

Having a medical alert button 2 (1.8) Not wearing loose clothing while cooking 2 (1.8)

Keeping a phone nearby 2 (1.8) Burning candles only in glass jars 2 (1.8) Avoiding stairs 2 (1.8) Checking furnace regularly 2 (1.8)

Trying to keep weight down 2 (1.8) Using lower heat on stove when cooking 2 (1.8)

Getting vision checked 1 (0.9) Monitoring use of space heater 1 (0.9) Other 10 (9.1) Other 12 (10.9)

1: The highlighted behaviors indicate Remembering When messages.

3. Remembering When™ Program Efficacy

3.a. Remembering WhenTM Safety Messages: Falls Prior to receiving the Remembering WhenTM program, most participants were engaged in activities recommended in the program, including exercising regularly (86%), taking their time to get up from sitting or lying down (83%), keeping stairs and walking areas free of clutter (73%), keeping a well-lit path between the bedroom and bathroom at night (73%), and using non-slip mats in the bathtub and on shower floors (66%) (Table 4). Following the Remembering WhenTM program, more participants cleared their paths of travel (n=104, 95%), compared to baseline (n=80, 73%) (p < 0.001). This improvement was noted for participants in both the home visit only study arm (baseline: n=46, 73%; follow-up: n=59, 94%) (p = 0.004) and the home visit and group presentation arm (baseline: n=34, 74%; follow-up: n=45, 98%) (p = 0.003). There was also a significant change in the percentage of participants using non-slip mats before (n=72, 66%) and after (n=86, 78%) receiving the Remembering WhenTM program (p = 0.003), which was also observed among participants in the home visit only study arm (baseline: n=41, 64%; follow-up: n=49, 77%) (p = 0.039). In addition, the program had an impact on the percentage of participants who turned on their lights before using the stairs (n=54, 49%), compared to the percentage turning on lights at baseline (n=51, 46%) (p = 0.035).

About half of the participants used grab bars on the walls next to the bathtub, shower and/or toilet (55%), used only throw rugs with a rubber, non-skid pad (49%) and turned on lights before using the stairs (46%) prior to receiving the Remembering WhenTM program (Table 4). Installing grab bars was the

—— Page 22 ——

only significant change following the program (baseline: n=60, 55%; follow-up: n=78, 71%) (p = 0.003), which was also observed among participants in the home visit and group presentation study arm (baseline: n=23, 50%; follow-up: n=33, 72%) (p = 0.021). In addition, only about one-quarter (24%) of the participants wore low-heeled shoes in their homes at baseline, and the program did not appear to have an effect on this behavior.

Although we observed significant improvements in some falls prevention behaviors between baseline and follow-up within each study arm, there was no difference in the percentage of participants improving their falls prevention practices between the two arms (Table 5). The exception was for keeping a well-lit path between the bedroom and the bathroom, where participants in both study arms experienced a decline in this behavior (home visit: 24%; home visit and group presentation: 17%). These declines were likely due to how participants understood the concept of “well-lit”. After receiving the home visit, and an in-depth home assessment by the fire department, they likely had a better understanding of how “well-lit” was defined within the context of the program.

3.b. Remembering WhenTM Safety Messages: Fires Most of the participants were engaged in fire safety practices prior to receiving the Remembering WhenTM program, including staying in the kitchen when frying foods (85%), having a telephone near their beds (79%), keeping space heaters at least 3 feet away from anything that can burn (75%), and understanding stop, drop and roll procedures if their clothes catch on fire (66%) (Table 6). There was a significant change in the percentage of participants understanding how to use stop, drop and roll procedures, including using a blanket or towel if they are unable to drop and roll on the floor, after participating in the Remembering WhenTM program (baseline: n=72, 66%; follow-up: n=93, 85%) (p < 0.001). At baseline, approximately half of the participants wore tight-fitting or short sleeves when cooking (52%), and this percentage did not change following the program.

Prior to receiving the Remembering WhenTM program, most participants had a smoke alarm installed outside each sleeping area (91%), on every level of their home (78%), and inside each bedroom (76%) (Table 6). Even with the high percentage of older adults engaging in smoke alarm safety practices at baseline, there was still a significant change in the percentage installing them on every level of the home (baseline: n=86, 78%; follow-up: n=107, 97%) (p < 0.001), which was observed in participants in both the home visit only study arm (baseline: n=52, 81%; follow-up: n=62, 97%) (p = 0.006) and home visit and group presentation study arm (baseline: n=34, 74%; follow-up: n=45, 98%) (p = 0.003). A significant change was also observed in the percentage of participants installing a smoke alarm inside each bedroom (baseline: n=83, 76%; follow-up: n=95, 86%) (p = 0.029). Few participants tested their smoke alarms monthly (11%), and there was no significant change in this behavior following the program.

There was no difference in the percentage of participants improving their fire safety practices between the two arms (Table 7).

—— Page 23 ——

Table 4. Percentage changes in Fall Safety Message recommendations between baseline and follow-up, overall and by study arm, n=110. Fall Safety Message

Total

Home Visit Only

Home Visit and Group Presentation

Baseline N (%)

Follow-Up N (%)

p-value1

Baseline N (%)

Follow-Up N (%)

p-value1

Baseline N (%)

Follow-Up N (%)

p-value1

Exercise Regularly Exercise at least 3 times a week

69 (86.3)

95 (90.5)

0.763

42 (85.7)

52 (88.1)

1.000

27 (87.1)

43 (93.5)

0.655

Take Your Time Take all the time you need to get up from sitting or lying down without rushing

90 (82.6)

98 (89.1)

0.189

52 (81.3)

55 (85.9)

0.518

38 (84.4)

43 (93.5)

0.289

Keep Stairs and Walking Areas Clear Keep stairs and walking areas free of electrical cords, shoes, clothing, books, magazines and other items

80 (73.4)

104 (95.4)

< 0.001

46 (73.0)

59 (93.7)

0.004

34 (73.9)

45 (97.8)

0.003

Improve the Lighting in and Outside Your Home a. Keep a well-lit path between your

bedroom and the bathroom b. See an eye specialist once a year2

79 (72.5)

72 (65.5)

0.337

50 (79.4)

39 (60.9)

0.019

29 (63.0)

33 (71.7)

0.252

Use Non-Slip Mats a. Use non-slip mats in the bathtub

and on shower floors b. Have grab bars installed on the wall

next to the bathtub, shower, toilet

72 (65.5)

60 (54.5)

86 (78.2)

78 (70.9)

0.003

0.003

41 (64.1)

37 (57.8)

49 (76.6)

45 (70.3)

0.039

0.096

31 (67.4)

23 (50.0)

37 (80.4)

33 (71.7)

0.070

0.021

Be Aware of Uneven Surfaces Use only throw rugs that have rubber, non-skid rug pads

54 (49.1)

62 (56.4)

0.215

30 (46.9)

33 (51.6)

0.648

24 (52.2)

29 (63.0)

0.133

Stairways Should be Well Lit a. Turn on lights before using stairs b. Have sturdy handrails on both

sides of the stairs

51 (46.4) 22 (27.8)

54 (49.1) 33 (41.8)

0.035 0.013

30 (46.9) 11 (24.4)

31 (48.4) 17 (38.6)

1.000 0.109

21 (45.7) 11 (31.4)

23 (50.0) 16 (45.7)

0.791 0.125

Wear Sturdy, Well-Fitting Shoes Wear low-heeled shoes

26 (23.6)

23 (20.9)

0.664

17 (26.6)

13 (20.3)

0.424

9 (19.6)

10 (21.7)

1.000

1: NcNemar test, comparing baseline and follow-up; 2: Unable to examine due to length of study follow-up (3 months) and recommendation being an annual behavior.

—— Page 24 ——

Table 5. Percentage improvement in Fall Safety Message recommendations across study arms, n=110. Home Visit

Only N (%)

Home Visit and Group Presentation

N (%)

p-value1 Exercise Regularly Exercise at least 3 times a week Improved Stayed the same Declined

3 (4.8) 54 (85.7)

6 (9.5)

3 (6.5) 41 (89.1)

2 (4.3)

0.696

Take Your Time Take all the time you need to get up from sitting or lying down without rushing Improved Stayed the same Declined

8 (12.5) 51 (79.7)

5 (7.8)

6 (13.3) 37 (82.2)

2 (4.4)

0.898

Keep Stairs and Walking Areas Clear Keep stairs and walking areas free of electrical cords, shoes, clothing, books, magazines and other treasures Improved Stayed the same Declined

16 (25.4) 44 (69.8)

3 (4.8)

12 (26.1) 33 (71.7)

1 (2.2)

0.897

Improve the Lighting in and Outside Your Home Keep a well-lit path between your bedroom and the bathroom Improved Stayed the same Declined

4 (6.3) 44 (69.8) 15 (23.8)

12 (26.1) 26 (56.5) 8 (17.4)

0.0062

Use Non-Slip Mats Use non-slip mats in the bathtub and on shower floors Improved Stayed the same Declined Have grab bars installed on the wall next to the bathtub, shower and toilet Improved Stayed the same Declined

10 (15.6) 52 (81.3)

2 (3.1)

13 (20.3) 46 (71.9) 5 (20.3)

7 (15.2) 38 (82.6)

1 (2.2)

13 (28.3) 30 (65.2)

3 (6.5)

0.954

0.333

Be Aware of Uneven Surfaces Use only throw rugs that have rubber, non-skid rug pads Improved Stayed the same Declined

11 (17.2) 45 (70.3) 8 (12.5)

9 (19.6) 33 (71.7)

4 (8.7)

0.750

—— Page 25 ——

Stairways Should be Well Lit Turn on the lights before using the stairs Improved Stayed the same Declined Have sturdy handrails on both sides of the stairs Improved Stayed the same Declined

7 (10.9) 51 (79.7)

6 (9.4)

3 (8.8) 29 (85.3)

2 (5.9)

8 (17.4) 32 (69.6) 6 (17.4)

4 (16.7) 17 (70.8) 3 (12.5)

0.331

0.4322

Wear Sturdy, Well-Fitting Shoes Wear low-heeled shoes Improved Stayed the same Declined

5 (7.8) 50 (78.1) 9 (14.1)

4 (8.7) 39 (84.8)

3 (6.5)

1.0002

1: Chi-square test, comparing improvement in falls prevention behaviors across study arms; 2: Exact test comparing improvement in falls prevention behaviors across study arms

—— Page 26 ——

Table 6. Percentage changes in Fire Safety Message recommendations between and follow-up, overall and by study arm, n=110. Fire Safety Message

Total

Home Visit Only

Home Visit and Group Presentation

Baseline N (%)

Follow-Up N (%)

p-value1

Baseline N (%)

Follow-Up N (%)

p-value1

Baseline N (%)

Follow-Up N (%)

p-value1

If You Smoke, Smoke Outside2 Provide smokers with large, deep, sturdy ashtrays. Wet cigarette butts and ashes before throwing them out or bury them in sand. Never smoke if medical oxygen is used in home.

Give Space Heaters Space a. Keep space heaters at least 3

feet away from anything that can burn- including you.

b. Turn off space heaters when you leave your home or go to bed.

21 (75.0)

17 (60.7)

20 (71.4)

15 (53.6)

1.000

0.754

13 (76.5)

10 (58.8)

12 (70.6)

9 (52.9)

1.000

1.000

8 (72.7)

7 (63.6)

8 (72.7)

6 (54.6)

1.000

1.000

Stay in the Kitchen When Frying Food a. Wear tight-fitting or short

sleeves when cooking. b. Stay in the kitchen when you fry.

52 (52.0)

82 (74.5)

51 (51.0)

88 (80.0)

1.000

0.238

30 (50.0)

51 (79.7)

30 (50.0)

51 (79.7)

1.000

1.000

22 (55.0)

31 (67.4)

21 (52.5)

37 (80.4)

1.000

0.031 If Your Clothes Catch Fire: Stop, Drop and Roll

72 (66.1)

93 (84.5)

< 0.001

41 (65.1)

50 (78.1)

0.049

31 (67.4)

43 (93.5)

< 0.001

Smoke Alarms Save Lives a. Have smoke alarms installed on

every level of your home. b. Have smoke alarms installed

inside each bedroom c. Have smoke alarms outside each

sleeping area d. Test alarms monthly

86 (78.2)

83 (75.5)

100 (90.9)

12 (11.1)

107 (97.3)

95 (86.4)

99 (90.0)

11 (10.1)

< 0.001

0.029

1.000

1.000

52 (81.3)

48 (75.0)

58 (90.6)

7 (11.1)

62 (96.9)

56 (87.5)

55 (85.9)

8 (12.7)

0.006

0.057

0.549

1.000

34 (73.9)

35 (76.1)

42 (91.3)

5 (11.1)

45 (97.8)

39 (84.8)

44 (95.7)

3 (6.5)

0.003

0.388

0.388

0.727

Plan and Practice Your Escape from Fire and Smoke3 Plan two ways out of every room in your home and two ways out of your home.

—— Page 27 ——

Know Your Local Emergency Number4 Your emergency number may be 9-1-1 or the fire department’s phone number.

Plan Your Escape Around Your Abilities a. Have a landline telephone or cell

phone and charger near your bed

b. Consider subscribing to a medical alert system

87 (79.1)

28 (25.7)

92 (83.6)

28 (25.5)

0.267

1.000

52 (81.3)

16 (25.4)

54 (84.4)

16 (25.0)

0.727

1.000

38 (82.6)

12 (26.1)

38 (82.6)

12 (26.1)

0.375

1.000

1: NcNemar test, comparing baseline and follow-up; 2: Did not use statistics due to the small number of participants who smoked or had a household member who smoked; 3: Questions not asked at baseline or follow-up because fire escape plans will vary by household, and questions to capture uniform knowledge were not possible. 4: Question not asked at baseline or follow-up because fire departments participating in the study all used 9-1-1 as the emergency number.

—— Page 28 ——

Table 7. Percentage improvement in Fire Safety Message recommendations across study arms, n=110.

Home Visit N (%)

Home Visit and Group Presentation

N (%)

p-value1 Give Space Heaters Space Keep space heaters at least 3 feet away from anything that can burn- including you Improved Stayed the same Declined Turn off space heaters when you leave your home or go to bed Improved Stayed the same Declined

2 (11.8) 12 (70.6) 3 (17.6)

3 (18.8) 9 (56.3) 4 (25.0)

1 (9.1) 9 (81.8) 1 (9.1)

1 (9.1) 8 (72.7) 2 (18.2)

0.6972

1.0002

Stay in the Kitchen When Frying Food Wear tight-fitting or short sleeves when cooking Improved Stayed the same Declined Stay in the kitchen when you fry Improved Stayed the same Declined

10 (16.7) 40 (66.7) 10 (16.7)

6 (9.4) 52 (81.3)

6 (9.4)

7 (17.5) 25 (62.5) 8 (20.0)

6 (13.0) 40 (87.0)

0

0.895

0.543

If Your Clothes Catch Fire: Stop, Drop and Roll Improved Stayed the same Declined

13 (20.6) 46 (73.0)

4 (6.3)

12 (26.1) 34 (73.9)

0

0.504

Smoke Alarms Save Lives Have smoke alarms installed on every level of your home Improved Stayed the same Declined Have smoke alarms installed inside each bedroom Improved Stayed the same Declined Have smoke alarms outside each sleeping area Improved Stayed the same Declined

11 (17.2) 52 (81.3)

1 (1.6)

11 (17.2) 50 (78.1)

3 (4.7)

4 (6.3) 53 (82.8) 7 (10.9)

12 (26.1) 33 (71.7)

1 (2.2)

8 (17.4) 34 (73.9)

4 (8.7)

4 (8.7) 40 (87.0)

2 (4.3)

0.258

0.978

0.626

—— Page 29 ——

Test alarms monthly Improved Stayed the same Declined

5 (8.1)

53 (85.5) 4 (6.5)

3 (6.7)

37 (82.2) 5 (11.1)

1.0002

Plan Your Escape Around Your Abilities Have a landline telephone or cell phone and charger near your bed Improved Stayed the same Declined Consider subscribing to a medical alert system Improved Stayed the same Declined

5 (7.8) 56 (87.5)

3 (4.7)

1 (1.6) 61 (96.8)

1 (1.6)

4 (8.7) 41 (89.1)

1 (2.2)

2 (4.4) 42 (91.3)

2 (4.4)

1.0002

1.0002

1: Chi-square test, comparing improvement in falls prevention behaviors across study arms; 2: Exact test comparing improvement in falls prevention behaviors across study arms

3.c. Perceived Susceptibility, Severity, and Fear of Falls and Fire On average, the 110 participants reported very low levels of perceived susceptibility of experiencing a fall (0.85 indicating an average response between “not at all likely” and “a little likely”), perceived severity of a fall (1.56 indicating an average response between “a little serious” and “somewhat serious”), and fear of falling (0.99 indicating an average response of “a little fearful”) at baseline. The levels of these three types of perceptions about falls did not change significantly after participation in the program (Table 8).

For fire, a slight increase in perceived susceptibility (from 0.10 at baseline to 0.20 at follow-up), decrease in perceived severity (from 2.52 to 2.25, from average being mid-way between “somewhat” and “very serious” to being closer to “somewhat serious”), and decrease in fear (from 0.95 to 0.71) were shown. There were no differences between the home visit only study arm and home visit and group presentation study arm in changes in these perceptions.

3.d. Perceptions about Prevention Behaviors for Falls and Fire Perceived efficacy to prevent falls increased among the participants from an average of 2.56 at baseline to 2.75 (indicating an average between “somewhat” to “very confident”) at follow-up (p=0.05) (Table 8). Similarly, self-efficacy to engage in “safety precautions” to reduce the chances of falling increased from baseline to follow-up, with average responses of 2.97 and 3.12 at baseline and follow-up, respectively. Perceived control over engaging in preventive behaviors also increased from an average of 2.50 (between “to some extent” and “to a great extent”) to 3.18 (above “to a great extent”). No changes were observed for perceived benefits of engaging in and intention to engage in preventive behaviors, with average responses indicating “very likely” that they would engage in preventive behaviors and “very likely” that engaging in these behaviors would help reduce the chances of a fall.

Perceived control over engaging in fire prevention behaviors also increased from an average of 2.44 (between “to some extent” and “to a great extent”) to 3.31 (between “to a great extent” and “to a very

—— Page 30 ——

great extent”). A change in self-efficacy to engage in fire preventive behaviors was observed, with an average indicating “very confident” that they can engage in fire prevention behaviors. This approached, but did not reach, statistical significance. There were no changes seen for self-efficacy to prevent a house fire or perceived benefits in engaging in, or intention to engage in, fire prevention behaviors. In general, participants felt “very confident” that they could prevent a house fire, “very likely” that engaging in preventive behaviors would help reduce their risks of experiencing fire, and “very likely” that they would engage in such preventive behaviors. There were no differences in changes in these perceptions between the two study arms.

3.e. Perceptions about Social Support from Others An increase in perceived support was reported in terms of information about falls prevention (from 1.86, “to some extent” at baseline to 3.00, “to a great extent” at follow-up), information about fire prevention (from 1.89 “to some extent” to 2.98, “to a great extent”) (Table 9). A similar increase was also reported for general social support perceptions such as emotional support, instrumental support, and support when sick or injured. No significant changes were seen for perceived social support participants received from others in reducing a fall or house fire. There were no group differences in the extent of changes in these social support perceptions between the home visit only and home visit and group presentation study arms.

—— Page 31 ——

Table 8. Perceptions about falls/fire and falls/fire prevention behaviors, n=110. Study Arm 1:

home visit only (n=64) Study Arm 2:

home and group (n=46)

All in Main Analyses (n=110) Group

difference Perceptions Baseline Follow-up Change p* Baseline Follow-up Change p* Baseline Follow-up Change p* p**

Perceptions about falls

Susceptibility 0.91 0.81 -0.10 0.359 0.78 0.87 0.09 0.471 0.85 0.84 -0.01 0.815 0.252 Severity 1.60 1.70 0.10 0.471 1.49 1.78 0.29 0.091 1.56 1.73 0.17 0.092 0.358 Fear 1.06 1.11 0.05 0.594 0.89 0.72 -0.17 0.132 0.99 0.95 -0.04 0.518 0.120 Efficacy to prevent 2.50 2.63 0.13 0.289 2.65 2.91 0.26 0.077 2.56 2.75 0.19 0.047 0.462

Efficacy: engage in behavior 2.95 3.16 0.21 0.022 3.00 3.07 0.07 0.569 2.97 3.12 0.15 0.038 0.329

Benefit 3.03 3.02 -0.01 0.902 2.91 3.09 0.18 0.221 2.98 3.05 0.07 0.501 0.324 Intention 3.17 3.27 0.10 0.400 3.11 3.28 0.17 0.198 3.15 3.28 0.13 0.136 0.652 Control 2.49 3.16 0.67 0.004 2.51 3.20 0.69 0.025 2.50 3.18 0.68 0.000 0.963

Perceptions about fire Susceptibility 0.11 0.22 0.11 0.070 0.09 0.17 0.08 0.160 0.10 0.20 0.10 0.021 0.797 Severity 2.52 2.22 -0.30 0.079 2.52 2.28 -0.24 0.175 2.52 2.25 -0.27 0.025 0.801 Fear 1.03 0.83 -0.20 0.074 0.85 0.54 -0.31 0.123 0.95 0.71 -0.24 0.019 0.652 Efficacy: prevent 2.91 2.92 0.01 0.888 2.85 3.02 0.17 0.132 2.88 2.96 0.08 0.307 0.330 Efficacy: engage in behavior 2.97 3.16 0.19 0.064 3.00 3.07 0.07 0.497 2.98 3.12 0.14 0.054 0.384

Benefit 3.05 3.11 0.06 0.531 2.87 3.11 0.24 0.147 2.97 3.11 0.14 0.128 0.354 Intention 3.22 3.21 -0.01 0.888 3.09 3.22 0.13 0.294 3.17 3.21 0.04 0.581 0.386 Control 2.48 3.31 0.83 0.001 2.40 3.31 0.91 0.002 2.44 3.31 0.87 0.000 0.839 Note: *p-values for differences between baseline and follow-up were derived via paired sample t-tests; **p-values for Group Differences were derived by conducting independent sample t-tests to compare the change scores (follow-up minus baseline) and comparing between Home visit only vs. home visit & group.

—— Page 32 ——

Table 9. Social support to reduce risks for fall/fire, information about prevention, and general emotional and instrumental support, n=110. Study Arm 1:

home visit only (n=64) Study Arm 2:

home and group (n=46)

All in Main Analyses (n=110) Group

differences Construct Baseline Follow-up Change p* Baseline Follow-up Change p* Baseline Follow-up Change p* p**

Support: to reduce fall 1.89 1.44 -0.45 0.058 1.87 1.85 -0.02 0.936 1.88 1.61 -0.27 0.132 0.235 Support: to reduce fire 1.72 1.83 0.11 0.668 1.85 1.93 0.08 0.754 1.77 1.87 0.10 0.593 0.953 Informational support: falls 1.80 2.97 1.17 0.000 1.96 3.04 1.08 0.000 1.86 3.00 1.14 0.000 0.788 Informational support: fire 1.83 2.95 1.12 0.000 1.98 3.02 1.04 0.000 1.89 2.98 1.09 0.000 0.776 Care when sick/injured 1.77 2.92 1.15 0.000 1.80 3.11 1.31 0.000 1.78 3.00 1.22 0.000 0.616 Emotional support 1.76 3.08 1.32 0.000 1.65 3.13 1.48 0.000 1.72 3.10 1.38 0.000 0.585 Instrumental support 1.73 3.20 1.47 0.000 1.65 3.17 1.52 0.000 1.70 3.19 1.49 0.000 0.845 Note: *p-values for differences between baseline and follow-up were derived via paired sample t-tests; **p-values for Group Differences were derived by conducting independent sample t-tests to compare the change scores (follow-up minus baseline) and comparing between Home visit only vs. home visit & group.

—— Page 33 ——

3.f. Discussion about Falls and Fire Prevention with Others The number of participants who reported discussing falls and fire prevention with their family and friends increased from baseline to follow-up in both study arms. At baseline, 34% had discussed falls prevention with family and friends, whereas 60% reported discussing this at follow-up (Table 10). Similarly, 21% had discussed fire prevention with family and friends at baseline, while 55% reported discussing this with family and friends at follow-up. The report of discussing fire prevention with health care providers also increased from 7% at baseline to 15% at follow-up. No changes were observed for discussing falls prevention with health care providers or senior center staff.

Table 10. Numbers who reported discussing about falls or fire with others at baseline and follow-up Discussed with…

Total

Home Visit Only

Home Visit and Group Presentation

Baseline N (%)

Follow-Up N (%)

p * Baseline N (%)

Follow-Up N (%)

p *

Baseline N (%)

Follow-Up N (%)

p *

Family member or friend: Fall

37 (34.0)

65 (60.0)

<0.001 20 (31.2)

37 (57.8)

<0.001 17 (27. 8)

28 (62. 2)

0.019

Family member or friend: Fire

23 (20.9)

61 (55.5)

<0.001 10 (15.6)

35 (54.7)

<0.001 13 (28.3)

26 (56.5)

0.004

Health care provider: Fall

34 (31.5)

31 (28.7)

0.728 22 (35.5)

17 (27.4)

0.332 12 (26.1)

14 (30.4)

0.803

Health care provider: Fire

8 (7.3)

16 (14.7)

0.039 5 (7.8)

10 (15.6)

0.125 3 (6. 7)

6 (13.3)

0.375

Senior center: Falling or Fire

22 (20.6)

25 (24.8)

0.424 10 (16. 7)

18 (30.0)

0.077 10 (24.4)

7 (17.1)

0.507

*P-values calculated using McNemar's exact test

4. Program Satisfaction and Suggestions for Improvement

4.a. Older Adult Participant Feedback At the end of the follow-up interview, many older adult participants made comments about their participation in the study and in Remembering When™. Of the 115 follow-up interview audio files available for analysis, 100 contained audio recordings of conversations between older adult participants and interviewers after the conclusion of the interview. The remaining 15 recordings ended at the conclusion of the interview and did not record subsequent conversations. In all, 55 participants gave feedback on the study and the Remembering When™ program and two offered suggestions for improvement. Most (n=47) of those participants explicitly expressed general satisfaction with their participation in the study and Remembering When™, saying, for instance, “I’ve enjoyed this” and “it was fun.” Nine participants specifically mentioned that they appreciated interacting with and learning new information from the Outreach Team members who conducted the group presentations and home visits. As one participant stated, “I really enjoyed the visit from the fireman… I thought it was very useful and he’s a lot of fun, too.” Several participants implied or stated that they already knew much of the information in the Remembering When™ program, but most were glad to have renewed awareness of falls and fire hazards: “it really reminded us of some things that we need to do in the home.” Other

—— Page 34 ——

themes included general satisfaction with interactions with the research team, learning something new from the program, hopefulness that their participation would benefit others, and increased confidence in their ability to make their home safe. Only three participants mentioned the monetary incentive, which two appreciated and one said was unnecessary. Suggestions for improving the participation experience for older adults were to re-word some of the confusing interview questions and also to include instruction on fire extinguishers in the program.

4.b. Outreach Team Member Feedback

Remembering When™ program feedback Overall, members of the Outreach Teams enrolled in the study expressed satisfaction with the Remembering When™ program. Specifically, respondents cited ease of use and its focus on relevant safety issues for their communities, as well as overall enjoyment using the program. As one respondent explained, “…it was just very well put together and it seemed to be—I guess I’m tryin’ to think of the word—very prepared. There wasn’t a lot of things that, other than scheduling the visits and everything, there really wasn’t much to do on our part, which was kinda nice. We just go and present the program as is and we didn’t have to adjust it or anything like that.” In addition to global feedback, Outreach Team members provided insights into how the program fits into their organizations’ work and culture, their views on the benefits and drawbacks of home visits and group presentations, the value of interacting with older adults in their community through the program, and recommendations to improve Remembering When™.

Alignment of Remembering When™ with fire department goals A prominent theme that emerged from the interviews was that the Remembering When™ program fits well with the outreach activities and goals of the participating fire departments. Many Outreach Team members valued the program’s potential to reduce the number of calls they receive from older adults, particularly for falls. One respondent noted that the majority of their calls, often ten per day, were older adult medical calls and stated, “if we can cut it down to nine or eight calls a day it would slow our call volume down without using all our resources.” For many Outreach Team members, doing something to prevent older adult incidents was a primary motivation to participate in the study.

Some respondents also said that Remembering When™ fills a gap in their fire department’s outreach strategy, which is often focused more on education for children or on general events that may be difficult for older adults to access:

The other thing that I like is that I don't feel like we do enough in fire prevention for seniors. They're not gonna come to the fire department Open House cuz they're not gonna drive in the winter. We don't do visits like we should. We do school visits, and everybody remembers the fire trucks. There really isn't a very good mechanism to reach seniors, and I think this program does that.

Perhaps related to this gap in older adult outreach is the difficulty in getting smoke alarm programs to reach older adults. Indeed, one Outreach Team member noted that the Remembering When™ home

—— Page 35 ——

visits allowed his team to not only increase awareness of their smoke alarm program, but also perform several installations at one time when neighbors stopped by and requested assistance during the visits.

Even other members of the fire departments who were not enrolled in the study appeared to support the program. Three Outreach Team members expressed that their coworkers were verbally supportive of their involvement in the Remembering When™ program and study, but that they did not get actively involved. Others noted that their colleagues accompanied them at times to group presentations and home visits, and that this experience fostered enthusiasm for the program:

We actually took some firefighters with us when we went into the houses doing it. We never really like to go in alone, just in case something ever happened. They all thought it was great, they wanna move forward with it...

Overall, respondents indicated that the Remembering When™ program was congruent with the needs and goals of their fire departments.

Home visit and group presentation feedback: Advantages and challenges When asked for feedback on the Remembering When™ group presentations and home visits, Outreach Team members identified strengths and weaknesses specific to each (Table 11). Some of the major benefits associated with the group presentations related to the interactions among older adults at the presentation, which fostered social connection and gave each older adult the benefit of hearing answers to each other’s questions. When Outreach Teams incorporated discussion and activities they were able to foster active engagement in the program material. Furthermore, some respondents indicated that the group presentation prepared the older adults to see the potential benefit of a home visit and made them feel comfortable with the presenter who would later come to their home:

They were glad that they came, they truly were… they really didn’t want us coming into their homes before then but we made them feel comfortable enough I think at that group presentation where it made it comfortable enough and they were okay with us coming… They kind of saw that if we come into their homes we might spot something that they missed.

On the other hand, the primary challenge that Outreach Team members faced with conducting group presentations was working with technology to show the PowerPoint presentation and videos. Overall, Outreach Team members expressed satisfaction with the group presentation format and valued its ability to educate many older adults at one time and foster positive interpersonal interactions.

Outreach Team members saw many advantages to the Remembering When™ home visits. Some found that the home visit provided an opportunity to correct misperceptions about what is or is not a hazard in the home and to provide “hands-on” tutorials for skills such as testing smoke alarms. Some Outreach Team members valued the one-on-one format, which allowed them to get to know the older adults and to see “that they were being attentive in the group presentations and had already made changes at their home.” There was a general sense from the respondents that the home visit seemed to be more “effective” than the group presentation in terms of reducing fall and fire hazards.

—— Page 36 ——

A couple of Outreach Team members had to overcome hesitation from older adults about inviting them into their homes:

…they were a little bit skeptical at first about us coming into their homes and trying to tell them what to do but we just went there and, “Today we’re not here to judge you or we can’t force you. We’re just here to help you try to make things better. Keep you in your home as long as we can.” They realized that so they were good with it.

As this respondent illustrated, Outreach Team members who met with this challenge of participants feeling uneasy about the home visit attempted to counter it by reassuring older adults that they would not use the home visit to criticize the person or his or her living conditions. Missed or forgotten appointments were another challenge for Outreach Teams; hence, they suggested doing reminder calls to each older adult shortly before the home visit.

Table 11. Advantages and challenges of group presentations and home visits mentioned by Outreach Team members in exit interviews. Advantages Challenges Group Presentations

• Opportunity for older adults to socialize and find common ground

• Opportunity for questions to get asked and answered that each participant otherwise may not bring up

• Outreach Teams working to foster active participation resulted in more engagement

• Preparation for the home visit, both in terms of behavior change and also increasing older adults’ comfort with having someone come to their home

• Difficulties with technology at various group presentation sites (e.g., setting up the PowerPoint presentation)

• Older adult shyness about asking questions in a group setting*

• Not enough time to cover all of the material*

• Discomfort for some Outreach Team members who do not like public speaking*

• Hard-to-find meeting rooms and locations*

Home Visits • Opportunity for Outreach Teams to establish personal connection with the older adult

• Opportunity for Outreach Team members to correct misperceptions about what is or is not a hazard in the home, address issues directly

• Home visits may be more effective than group presentations at changing behavior/ conveying information

• Opportunity to get feedback on the effectiveness of the group presentation by seeing what changes were made*

• Older adult nervousness about having someone come to their home to do a home visit

• Missed home visit appointments because older adults forgot – need to remind them

*Mentioned by only one participant.

—— Page 37 ——

Interactions with older adults Interacting with older adult participants was a particularly rewarding aspect of Remembering When™ program delivery for many Outreach Team members. Several respondents mentioned that they gained a new understanding of the interests, concerns, needs, and assets of older adults. Concern about safety in the home and a desire for independence were primary motivators for the older adult participants that Outreach Team members discovered. They also highly valued getting to know these older members of their communities:

I think as a whole it helped me to reconnect with the human element of people. Because it's easy when you're in a police car or a fire truck drivin' by or goin' to a call to not think about the house you're drivin' by.

For this Outreach Team member and others, connecting with the older adult participants was rewarding in and of itself, and Remembering When™ was the driver of those interactions. Some also noted that the older adult participants seemed to respect the fire department officials more than others do. Of note, a couple of respondents emphasized the importance of having the skills to communicate effectively with older adults so that they feel comfortable.

Many Outreach Team members also learned the degree to which the fire and fall prevention needs of older adults can vary, even within one community. Many older adults who were recruited to participate were already knowledgeable of fire and fall prevention practices:

… most of them were really on top of things, I mean they really were. We nitpicked the residents as much as we could. We found two or three things that could be changed but like I said if you can change just one two things that really makes my day.

As this respondent indicated, the Outreach Teams were often still successful in helping even very knowledgeable older adults to improve safety. Some of this variation in needs was also due to the living situation of the older adult. Those living in senior housing facilities already had many recommended practices in place due to regulations, so Outreach Team members took less time with those home visits than with visits in independent residences where there were often more hazards to address. Some respondents found, however, that older adults living sometimes brought up “sensitive issues” about their senior facilities, such as maintenance oversights, and reported these complaints to the Outreach Teams. When dealing with lower-income older adults, some Outreach Team members wished they had information available on low-cost options for improving safety to give them. These respondents sometimes felt at a loss to help, asking, for instance, “Is there someone I can send them that will do some electrical work for free?” As with this respondent, some Outreach Team members questioned the degree to which they could tailor the Remembering When™ program information to suit the needs of each older adult. There was a sense that such tailoring was important for making the program effective: “I think giving the presenter the openness to be able to present what the person is more interested in is a great thing.” Overall, the need for adapting Remembering When™ program delivery based on the needs of the older adults was a central theme.

—— Page 38 ——

Recommendations for the Remembering When™ program When asked what they would change to make the group presentations and home visits more effective, most Outreach Team members offered logistical suggestions (Table 12). For instance, one of the primary concerns regarding the group presentations was ensuring that the PowerPoint presentation was able to run smoothly. Thus, some respondents recommended having better technical support or logistical considerations for group presentations. For the home visits, one concern was that older adults sometimes forgot about their appointments and were either not at home or unprepared for company. The recommendation, therefore, was to make reminder calls a part of the protocol. Two Outreach Team members who had carried out many of the program delivery activities by themselves recommended that at least two people conduct each home visit, both to verify the home hazard check and to help write up an action plan for the older adult.

Table 12. Recommendations for the Remembering When™ group presentation and home visit mentioned by Outreach Team members in exit interviews. Group Presentation Home Visit Arrange for better technical support for PowerPoint presentations and multimedia at each group presentation venue

Call home visit participants to remind them of their appointment just prior to the home visit

Select locations that are easy to find* Have at least two Outreach Team members do each home visit

*Mentioned by only one participant.

For the overall program delivery strategy, the predominant opinion among respondents was that Remembering When™ would be most effective as a group presentation followed by a home visit for those who desire it. The potential benefit of this approach, in the view of some respondents, would be to use the group presentation to first “gain [older adults’] trust to get in their home” for the home visit. Outreach Team members sometimes viewed the home visit as more effective than the group presentation. As one respondent stated, “I really think that gettin’ home visits is where it’s at, that’s the meat and potatoes of the program, where we just scratch the surface with the group presentations.” Conducting the home visit after the group presentation also gives Outreach Teams an opportunity to see any changes that older adults made in response to the group presentation material. To avoid redundancy and improve efficiency, one respondent recommended giving the instructional material primarily during the group presentation:

Maybe what I’d do would be a group presentation and then if they really continued to be interested or want to pursue and make sure they’re really safe, offer a home visit on top of that. Then you might even be able to cut short the home visit a little bit more because you don’t have to go through the presentation again.

In addition to the group presentation and home visit, two Outreach Team members suggested that older adults would be more likely to make and maintain safety behavior change if they knew there would be a follow-up check from the fire department a few months or a year after the first home visit. These respondents suggested adding such a follow-up home visit to the program.

—— Page 39 ——

Pilot evaluation study feedback Overall, most Outreach Team members expressed general satisfaction with their participation in the study. The main barrier to their satisfaction, and to accomplishing study activities, was having to schedule and arrange the home visits, a task for which they had insufficient time or staffing. One Outreach Team member summarized these points by saying, “Outside of the scheduling aspect of it, I can’t think of anything that was un-pleasurable.” Feedback from respondents on each part of their pilot evaluation study experience is summarized in the following sections.

Initial Outreach Team training The Outreach Team members were generally satisfied with the Remembering When™ program training and study protocol training they received prior to the start of the study. They appreciated learning “how to deliver, how to properly use the handouts, how to use the flipcharts” and receiving the program materials. Although feedback was generally positive, some stated that they would have liked to receive more in-depth training on how to conduct home visits and which hazards to look for. One Outreach Team member also expressed a desire for more guidance on how to communicate with older adults during a home visit, saying, “For me, it’s like okay, when I go into this home, what precisely do you want me to explain to them about? What is unsafe and what is safe? How do I tactfully talk about clutter…?” Additionally, two respondents wanted more information on how to access NFPA information resources and databases containing facts for responding to older adult questions. A broader recommendation mentioned by one Outreach Team was to move the training closer to the start of program delivery since “there was a lot of time to just forget about the information” in the intervening period.

Older adult recruitment Frustration at low older adult recruitment was widespread among Outreach Team members. As one respondent explained, “I think it’s a vast need, and I think well, it’s a drop in the bucket of what we’ve done, so I think that it definitely needs to be continued.” Others echoed this theme of a great need for fall and fire prevention activities among older adults coupled with dissatisfaction that so few signed up to participate in the study. Some respondents believed that older adults may have been reluctant to sign up for the study because of the home visit component:

The biggest thing is people they’re a little skeptical about us coming into their homes… I think that’s the trouble with people wanting to sign up for it. I don’t know if people want us coming into their homes and trying to tell them what to do. Somehow we have to convey that we’re not there to judge them, gain their trust I guess.

As mentioned previously, one recommendation for making older adults feel more comfortable about the home visit was to deliver the program as a group presentation, first to familiarize the older adults with the presenters. To boost recruitment, recommendations included speaking to local community groups such as the Kiwanis Club and having older adults who have been through the program help at recruitment events, perhaps even giving videotaped testimonials. When asked specifically about recruiting homebound older adults, many Outreach Team members agreed that it is an important group to reach. Some suggested connecting with local, trusted healthcare providers, visiting nurses, and other health-related or home-visiting organizations to recruit homebound older adults. In general, many

—— Page 40 ——

Outreach Team members agreed about the need to recruit more older adults into the study and felt that deeper penetration into the community may yield better results.

Outreach Team time and staffing limitations One of the major barriers Outreach Teams experienced in completing the study activities, which included Remembering When™ program delivery and completing data collection instruments, was limited time and staff support. One Outreach Team member described the issue as follows:

I mean I really enjoyed it, but it was we were trying to get a lot of visits in a short amount of time. For me, being really the only person doing them, I felt a little scatterbrained [laughter] or a little hectic some days trying to make sure I was getting everything done.

Some “teams” operated with only one active member who conducted all home visits and group presentations, whereas others had two people engaged. Even for some two-member teams, scheduling home visits and delivering them was difficult to do alongside other fire department duties. Lack of time and a staffing shortage contributed to one Outreach Team being unable to complete the study. A representative of that team stated that the study time commitment was not clear at the beginning and wished that the research team had given them a better idea of how much time would be required.

To overcome the time and staffing barrier, Outreach Team members made several recommendations. First, although it is not feasible in every fire department, some respondents recommended that Outreach Teams consist of at least two active members who can attend home visits together, with other members of the fire department trained in Remembering When™ and prepared to assist when needed. Second, some suggested that the research team could assist with scheduling since, as one respondent said, “I think that would have taken quite a bit of burden off of us, or a lot of work load off of us and we could have just been able to show up, deliver the presentation.” Finally, for some the study coincided with a difficult time in the fire department characterized by unusually low staffing levels and competing priorities. It may be difficult to anticipate these busy periods, but when possible, some Outreach Teams suggested finding a better time of year for the study, such as before summer and fall.

Data collection Most Outreach Team members were satisfied with the primary data collection instrument they used to record their group presentation and home visit experiences, the Program Delivery Log. Nearly all respondents reported that they completed the logs very soon after each home visit or group presentation, and that the form was “straightforward, pretty easy” to use. For most respondents, having the flexibility to complete the forms on paper was convenient, although one recommended having an electronic option for completing them as well. One concern that emerged from a few interviews was regarding the breadth of options and items listed on the Program Delivery Log, particularly for the home visit. Having so many items listed on the form led some respondents to second-guess their program delivery strategy and assume that they should have done more: “I saw a lotta things that I didn't check, and it kinda made me feel like I wasn't doin' as good a job as I should be doin'.” There were no specific recommendations for the content of the Program Delivery Log except that spaces should be added to

—— Page 41 ——

allow Outreach Teams to explain why some topics were not covered as much as others, often due to variations in older adult needs.

Incentives Opinions varied greatly on the monetary incentive offered to Outreach Team members participating in the study. One respondent said the amount was “adequate reimbursement” for the time and effort spent on study activities. On the other hand, another participant said, “for the amount of work load that went in, it was a very minimal amount of money” if it were offered to a volunteer or unpaid fire official. More respondents said that the incentive was “unnecessary” or even “inappropriate” because some Outreach Team members were paid through their departments for time spent on study-related activities. Among some, there was a sense that Outreach Teams “need to be doing this stuff anyway,” regardless of incentive. Instead of giving the incentive money to the Outreach Team member, one respondent suggested using those funds to provide items for Outreach Teams to distribute to low-income older adults, such as bath mats, night lights, or handrails. There was much greater consensus on the value of the $100 gift cards to purchase props for group presentations, with respondents noting that many fire departments may not have the funds to purchase them and that props and visual aids “are probably the best teaching tool.”

Information sharing among Outreach Team members One Outreach Team raised the idea of sharing best-practices for home visits and group presentations, troubleshooting strategies, and tips with other Outreach Team members on how to deliver the Remembering When™ program, noting that such networking is valuable:

Getting feedback or maybe sharing ideas back and forth with them maybe through the full process would have maybe helped a little bit….Yeah because then you maybe they had some ideas that we didn’t think of or vice versa. That’s one thing I’ve learned in the fire service every training session I go to I can learn more by sharing information with other departments.

They expressed a desire to meet with the other teams again about mid-way through the study participation period and also once the study activities concluded.

Next steps for the Outreach Teams Although the Outreach Team members were not asked whether and how they would continue delivering Remembering When™, some volunteered that they would strive to continue the program. Those aspiring to continue with Remembering When™ talked mainly about questions of how to recruit older adults, particularly those most at risk, to participate:

If there was a way that I could—I will find opportunities to deliver that program in the right environments. I hope what comes outta this a little bit of research into how do we get into those vulnerable populations that we don't have contact with?

To allow the program to continue, some respondents mentioned that they may decide to train others in their fire department in Remembering When™ so that they could “expand the program.” This

—— Page 42 ——

orientation toward sustained delivery of Remembering When™ in their communities appears to emerge naturally from respondents’ focus on the needs of older adults and their satisfaction with the program.

Discussion Our results suggest that an educational program, like Remembering WhenTM, may improve awareness of falls and fire prevention. When asked about the types of fall and fire safety behaviors in which participants were engaged, few offered behaviors mentioned in the program. However, when asked whether they were engaged in specific behaviors related to program messages, most of the participants indicated doing them. There appears a potential disconnect between activities older adults engage in and how they are associated with falls and fire prevention.

In terms of how program recommendations to reduce fall and home fire risks were translated to behavior change, we found that participants improved some falls and fire safety behaviors after participating in the Remembering WhenTM program. For falls, these behaviors included keeping paths clear of clutter, using non-slip mats in the bathtubs and shower floors, installing grab bars next to the bathtub, shower and toilet, and turning on lights before using the stairs. For home fire safety, the behaviors included understanding stop, drop and roll techniques and having a smoke alarm installed on every level of the home. Before participating in the program, many of the older adults were already engaged in several of the falls and fire safety behaviors recommended in the program. By approaching a population already engaged in falls and fire prevention, the program may benefit older adults by encouraging continued positive behaviors. However, for those not engaged, Remembering WhenTM teams can impact behavior change, which is suggested in our findings.

The Remembering When™ program aims to promote preventive behaviors by highlighting the risks associated with falls and home fires and providing recommendations on ways to reduce such risks. Our results suggest that the program was successful in increasing the perceptions of susceptibility to and severity of experiencing a home fire among the participants. This change was accompanied by participants’ increased perceptions of control over their fire prevention behaviors (decreased sense of barriers). An increase in self-efficacy to engage in these preventive behaviors approached, but did not reach, statistical significance. In terms of falls prevention, perceptions of risk did not change among the participants after the program. However, there was an increase in participants’ self-efficacy to prevent a fall and also to engage in falls prevention behaviors. Furthermore, participants’ sense of control over their preventive behaviors increased, or perceived barriers to engage in preventive behaviors decreased, after participating in the program. These findings suggest that the mechanisms of influence that the Remembering When™ program has on participants’ cognition may differ for fall and fire prevention components. For example, the reason changes in fire prevention behaviors were observed may be due to an increased perception of risk among the participants, whereas the reason for changes in falls prevention behaviors may be due to a perceived reduction in barriers.

There were no consistent findings to suggest that the Remembering WhenTM program was more efficacious in changing falls and fire prevention behaviors when delivered using the home visit only format, compared with the home visit format with a group presentation. While there were some

—— Page 43 ——

significant changes in behavior within each training format, when improvement proportions were compared across the study arms, there were no statistically significant changes. These findings suggest that the group presentation may not have an additional benefit to behavior change above what the home visits are already providing. This is consistent with the literature indicating that more tailored interventions (as would be provided in a home visit format) produce greater changes in health behaviors than non-tailored interventions (as would be provided in a group presentation) (Lustria et al., 2013; Noar et al., 2007). However, the Outreach Teams indicated that the group presentations helped prepare the older adults to understand the potential benefits of a home visit and made them feel more at ease with the fire fighter who would later come to their home. In this way, the group presentation may be valuable as a way to gain entry to the home visits. We should note, however, that the lack of statistical significance in these findings may be due to small sample sizes and/or how we chose to analyze the data. In the latter, rather than examining incremental changes in behavior, we dichotomized the behaviors as meeting the Remembering WhenTM recommendation or not meeting the Remembering WhenTM recommendation. This dichotomy offers less power to detect differences.

Participants reported increased informational support from the people around them about falls and fire prevention, likely due to becoming acquainted with the local fire fighters through the program. The observed increases in general social support, such as emotional and instrumental support, were not hypothesized at the beginning of the study. One possible explanation for this increase in social support is the positive social interactions participants enjoyed while participating in Remembering When™ and interacting with Outreach Team members and fellow participants. We hypothesized that the number of participants who discussed falls and fire prevention with others would increase after program participation, as sharing of prevention information by participants with others is one of the unintended consequences we often see when implementing public health programs. As expected, the percentage of participants who discussed falls and fire prevention with family and friends significantly increased after the program. The percentage who talked with health care professionals increased only for the topic of fire prevention. In further analyses, we may seek to understand whether discussing risks and preventive behaviors with family, friends and health care professionals is associated with an increase in perceived social support.

The project has limitations we need to address. First, study findings may not be generalizable to all older adult populations in the U.S. We conducted a small-scale randomized controlled trial of older adults living in Iowa. In addition to the small sample size, we did not sample older adults in a manner that would make them representative of a larger population. Second, we intended to include homebound older adults in the study, but the research and Outreach teams were unable to successfully reach them to describe the study and solicit interest in participating. The primary barrier from the home visiting organizations we approached was a concern about violating HIPAA laws. This barrier could have been addressed by having a representative from a home visiting organization partner with the fire departments on the Outreach Teams. Finally, our original design randomized participants into two intervention study arms: (1) those receiving the Remembering WhenTM program in home visit format only, and (2) those receiving the Remembering WhenTM program in both a group presentation and home visit format. However, not all participants received their allocated intervention (n=21). In response, we

—— Page 44 ——

re-allocated them into the more appropriate study arm and analyzed them based on the intervention they received rather than where they were randomized. Even though there were exceptions to randomization, we did not find any differences in demographics and risk factors for falls and fires between the two study arms. We, therefore, feel fairly confident that the randomization (even with exceptions) held.

In future analyses we plan to seek further understanding of the mechanisms through which behavioral change may have occurred through the mediating effects of changes in perceptions, social support and discussions about risks and preventive behaviors. Based on the behavioral change theories we considered in this study, we expect to see that high levels of perceived risks to falls and house fires (but not too high to elicit fear), accompanied by higher or increased levels of self-efficacy to engage in behaviors and decreased perceived barriers to engage in these behaviors, would lead to improved changes in falls and fire safety behaviors. Finally, future studies should consider an approach that measures how the Remembering WhenTM program is adapted to meet the individual needs of older adults and examine the effectiveness of this approach in changing falls and fire prevention behaviors. Another future study should examine recruitment strategies for reaching vulnerable, at-risk populations of older adults to better understand how Remembering WhenTM can be effective in increasing falls and fire safety behaviors and influencing related perceptions.

Recommendations Recommendations for delivering the Remembering WhenTM program and conducting future studies are provided below.

Older adult recruitment and retention. The most successful recruitment methods for the research team were in-person recruitment events, mailed information, and announcements on email listservs and other media. We found testimonials from older adults who were already participating in the program to be valuable in recruiting additional participants, as well as recruiting via trusted community members, such as healthcare providers. It was important to emphasize the connection between the Remembering When™ program and maintaining independence, which is a key concern for many older adults, and to assure older adults that the aim of the program was not to judge or criticize. Future studies should focus on recruiting populations with fewer Remembering When™ program recommendations in place. This would further support the effectiveness of the program in changing behaviors, in addition to potentially reaching populations more at risk for a fall or house fire. Finally, although participant retention in the study was high, we would recommend that more options for dates, times and locations for group presentations of the program be available.

Recruitment, composition and retention of Outreach Teams. We would recommend that Outreach Teams partner with a home visiting organization to improve recruitment of homebound older adults and those at high risk for a fall and/or house fire, as well as to improve the convenience of conducting home visits. We would also recommend that Outreach Team composition include at least two fire department members, committed to Remembering When™ program delivery, with the possibility of training more members from the department. Outreach Teams appreciated the funds provided to them to purchase

—— Page 45 ——

props for program delivery but had suggested more facilitation of information-sharing and networking among teams during the program delivery period. In terms of future research, we would recommend putting a research collaboration agreement or memorandum of understanding in place that is signed by the fire chief and the institution conducting the research to improve retention of Outreach Team members and follow-through.

Outreach Team training. A common training suggestion offered by the Outreach Teams was to provide more information about how to conduct the home visits, how to communicate tactfully and effectively with older adults, and what they should be specifically assessing during the home visit. They also wanted more tips on how to tailor the program to suit the diverse needs of older adults. We would recommend that the extent to which tailoring of the program components occurred and the perceptions of effectiveness of such tailoring be captured as part of the fidelity assessment. This would be helpful in future studies to better quantify implementation of program delivery. Finally, Outreach Teams felt that holding training closer to the start of program delivery was important.

Procedures. We have two primary procedural recommendations: (1) continually remind home visit only participants not to attend group presentations, and (2) provide assistance to Outreach Teams in scheduling home visits, whether by a research team member or dedicated staff member from the team’s fire department.

Remembering When™ program delivery in the community. Outreach Team members felt they needed more assistance in determining what additional information could be included with the Remembering When™ program, and what information could be removed, to flexibly adapt the program to the needs of the older adult participants. Many also thought that adding a follow-up check or booster home visit to give older adults some accountability to make the recommended changes, and to provide a reminder of the recommendations, would be important for sustained behavior change. Many also felt that having two Outreach Team members conduct each home visit to double-check the home hazard checklist and help write up an action plan would improve the comprehensiveness and accuracy of the recommendations. Finally, some recommended providing a list of local resources to which they could refer older adults for home safety improvements or discounted products.

In terms of how the program was delivered, there was a suggestion from the Outreach Teams that the program only be delivered as a group presentation, making a home visit available only to those who sign up or request it. Group presentations are better for reaching more older adults but do not offer the opportunity to make individualized recommendations, which Outreach Teams can do in a home visit. Many Outreach Team members preferred the home visit approach because they could better adapt the program to the needs of the older adults. There were comments about how it was important to reassure older adults that home visits are not meant to be about judging or criticizing them or their living conditions, and that conducting a group presentation first could be a way of gaining that trust.

—— Page 46 ——

Acknowledgments • Fire Protection Research Foundation • Karen Berard-Reed • Audrey Schroer • Talor Gray • Erin Robinson • Sara Cassabaum • Ellen Schafer • Fire Protection Research Foundation Technical Panel • “The Institute for Clinical and Translational Science at the University of Iowa is supported by the

National Institutes of Health (NIH) Clinical and Translational Science Award (CTSA) program, grant U54TR001013” [REDCap]

—— Page 47 ——

References AGS/BGS Panel on Prevention of Falls in Older Persons, American Geriatrics Society and British Geriatrics Society. (2011). Summary of the updated American Geriatrics Society/British Geriatrics Society (AGS/BGS) clinical practice guideline for prevention of falls in older persons. Journal of the American Geriatrics Society 59(1):148-157.

Callahan CM, Unverzagt FW, Hui SL, Perkins AJ, Hendrie HC. (2002). Six-item screener to identify cognitive impairment among potential subjects for clinical research. Medical Care 40(9):771-781.

Centers for Disease Control and Prevention. (2008). Self-reported falls and fall-related injuries among persons aged at least 65 years - United States, 2006. MMWR 57:225-229.

Champion VL, Skinner CS. The Health Belief Model. (2008). In: Glanz DK, Rimer BK, Viswanath K, eds. Health Behavior and Health Education: Theory, Research, and Practice. 4th ed. San Francisco, CA: Jossey-Bass:45-65.

Clare J, Garis L, Plecas D, Jennings C. (2012). Reduced frequency and severity of residential fires following delivery of fire prevention education by on-duty fire fighters: Cluster randomized controlled study. Journal of Safety Research, 43(2):123-128.

Flynn FD. Characteristics of home fire victims. Quincy, MA: National Fire Protection Association, Fire Analysis and Research Division, 2010.

Harris P, Taylor R, Thielke R, Payne J, Gonzalez N, Conde JG. (2009). Research electronic data capture (REDCap) - A metadata-driven methodology and workflow process for providing translational research informatics support. Journal of Biomedical Informatics 42(2):377-381.

Herr M, Latouche A, Ankri J. (2013). Homebound status increases death risk within two years in the elderly: results from a national longitudinal survey. Archives of Gerontology and Geriatrics 56(1):258-264.

Hill AM, Hill K, Brauer S, Olver D, Hoffmann T, Beer C, McPhail S, Haines TP. (2009). Evaluation of the effect of patient education on rates of falls in older hospital patients: description of a randomised controlled trial. BMC Geriatrics 9:14.

Jones J, Casteel C, Gildner P, Bowling JM, Blalock SJ. (2014). Falls prevention among ambulatory and homebound older adults. Presentation at the Aging in America Conference, American Society on Aging, San Diego, CA.

Lustria ML, Noar SM, Cortese J, Van Stee SK, Glueckauf RL, Lee J. (2013). A meta-analysis of web-delivered tailored health behavior change interventions. Journal of Health Communication 18(9):1039-1069.

Marshall SW, Runyan CW, Bangdiwala SI, Linzer MA, Sacks JJ, Butts JD. (1998). Fatal residential fires: who dies and who survives? JAMA 279(20):1633-1637.

—— Page 48 ——

National Center for Injury Prevention and Control. (2013). WISQARS Fatal and Nonfatal Injury Reporting System. Office of Statistics Programming, Centers for Disease Control and Prevention.

Noar SM, Benac CN, Harris MS. (2007). Does tailoring matter? Meta-analytic review of tailored print health Behavior change interventions. Psychological Bulletin 133(4):673-693.

Ornstein KA, Leff B, Covinsky KE, Ritchie CS, Federman AD, Roberts L, Kelley AS, Siu AL, Szanton SL. (2015). Epidemiology of the homebound population in the United States. JAMA Internal Medicine 175(7):1180-1186.

Ottoson J, Rivera M, Gordon C. (2004). Evaluation Report: Remembering When Fire and Fall Prevention Program for Older Adults. Report submitted to: Centers for Disease Control and Prevention.

Remembering When: A Fire and Fall Prevention Program for Older Adults. Downloadable at: http://www.nfpa.org/safety-information/for-public-educators/education-programs/remembering-when.

Robertson MC, Gillespie LD. (2013). Fall prevention in community-dwelling older adults. JAMA 309(13):1404-1407.

Rubenstein LZ, Vivrette R, Harker JO, Stevens JA, Kramer BJ. (2011). Validating an evidence-based, self-rated fall risk questionnaire (FRQ) for older adults. Journal of Safety Research 42(6):493-499.

Runyan CW, Johnson RM, Yang J, Waller AE, Perkis D, Marshall SW, Coyne-Beasley T, McGee KS. (2005). Risk and protective factors for fires, burns, and carbon monoxide poisoning in U.S. households. American Journal of Preventive Medicine 28(1):102-108.

Stubbs B, Brefka S, Denkinger MD. (2015). What works to prevent falls in community-dwelling older adults? An umbrella review of meta-analyses of randomized controlled trials. Physical Therapy 95(8):1-16.

Ta VM, Frattaroli S, Bergen G, Gielen AC. (2006). Evaluated community fire safety interventions in the United States: a review of current literature. Journal of Community Health 31(3):176-197.

Tinetti ME, Richman D, Powell L. (1990). Falls efficacy as a measure of fear of falling. Journal of Gerontology 45(6):P239-P243.

US Fire Administration / National Fire Data Center. (2006). Fire and the older adult. Washington, DC: US Department of Homeland Security.

US Fire Administration. (2015). Civilian fire injuries in residential buildings (2011-2013). Topical Fire Report Series 16(3):1-12.

van Belle G, Fisher LD, Heagerty PJ, Lumley T. (2004) Biostatistics: A Methodology for the Health Sciences, Second Edition. Hoboken, NJ: Wiley.

WHO. (2007). Global Report on Falls Prevention in Older Age. Geneva, Switzerland: World Health Organization.